Search Results (9,922)

The spread of unmanned aerial vehicles in daily operations makes the early and reliable diagnosis of propeller faults necessary. However, the performance values reported for such systems are usually obtained with sample level splits, and it is not known which feature representation should be matched with which learner. In this study, a leakage-free feature–model matching framework is presented for propeller fault classification. Microphone and six-axis inertial measurement unit data have been collected on a test bench with 980 kV and 1400 kV motors for one healthy and eight faulty propeller conditions at 16 throttle levels, and 8490 windows of 1 s have been extracted from 1735 measurement files. Four scalar feature sets and three time-frequency representations have been matched with seven ensemble learners and three compact convolutional networks under a file atomic split, and the permutation ranking of the best model has been returned to the feature selection stage. The highest macro-F1 value of 0.8027 and an accuracy of 0.8816 have been obtained with the stacked ensemble trained on the 52 input subset ranked by explainability. It is seen that the time domain statistics and the accelerometer axes are dominant, that three inertial axes reach a macro-F1 of 0.7661, and that cepstral and envelope features stay below the Welch-based features at the sampling rate of 90.9 Hz. The cross-motor experiments have shown that the models depend strongly on the motor class, and the McNemar test has confirmed that the difference between the ensemble branch and the compact convolutional branch is not accidental. In this way, the framework can be used as an evaluation protocol for low-cost multisensor setups on low-level devices. Full article

Environmental-monitoring systems often bundle sensing, communication, storage, visualization, and control into one application, making later changes difficult. We designed a service-oriented platform that separates these functions through defined interfaces. It combines a Raspberry Pi gateway, a dedicated motor-control microcontroller, five environmental sensor modules, Node-RED middleware, a database, and a web interface. Deterministic code alone evaluates threshold and composite rules and controls safety-relevant alerts; an optional large language model (LLM) turns pre-computed statistics and rule outcomes into narrative reports. We examined data acquisition and rule processing during two short indoor campaigns. In the residential campaign, the SCD41 yielded 78 valid three-minute bins (234 min of recorded data) across four sessions between 09:18 and 17:12 local time; binned CO₂ concentrations ranged from 679 to 1471 parts per million (ppm). Using the initial campaign for development and the residential campaign as a temporal holdout, the persistence model produced a 15 min forecast mean absolute error of 58.3 ppm and a root mean square error of 78.8 ppm. A separate controlled experiment generated 270 reports from nine deterministic synthetic scenarios. Every reporter preserved all deterministic alert identifiers, while the fixed template and seven of the nine locally hosted LLMs achieved 100% numerical fidelity. Qwen 3.5 9B was the only LLM that returned all required measured content without automated claim-review flags and produced identical outputs across repetitions for every scenario. These results confirm integration and functional separation under the tested conditions, but they do not demonstrate week-scale reliability, longer-horizon forecasting accuracy, robotic mobility performance, or load scalability. Full article

During the development of modern electric drives, vibroacoustic quality is one of the most critical parameters alongside motor performance and efficiency. One source of poor vibroacoustic characteristics is the inhomogeneity and imperfection of the soft magnetic materials found in the electrical machine. The primary indicators of these defects are the selected electrical steel and the machining processes used for the stator (waterjet cutting, laser cutting, punching). These inhomogeneities excite the mass-spring-damper oscillating system through higher-order harmonics induced in the air gap. The objective of the present research is the identification of these air-gap harmonics, for which this study aims to perform a systematic investigation chain starting from closed magnetic circuit measurements and building up to measurements loaded with an air gap. This is based on classical Epstein measurements, followed by induction measurements conducted in the air gap of a C-core. The test specimens were fabricated from M400-50A electrical steel using three different machining methods. The results show that the machining process has a significant impact on the magnetic characteristics, and the exit angle of the induction vectors measurable in the air gap also deviates from the ideal. Full article

To improve the dc-bus voltage utilization while reducing the commutation burden of an insulated gate bipolar transistor (IGBT)-based open-end winding permanent magnet synchronous motor (OEW-PMSM) drive, this paper proposes an improved alternate sub-hexagonal center PWM (ASHCPWM) strategy. The drive employs two three-phase IGBT inverters sharing a common dc bus. Since the turn-off energy of an IGBT is strongly affected by its switching frequency, ASHCPWM alternately clamps one inverter and reduces the number of switching transitions. The resulting common-mode voltage mismatch, together with the dead-time voltage error required to prevent shoot-through in the IGBT bridge legs, nevertheless produces significant zero-sequence current and additional semiconductor current stress. A mathematical model of the zero-sequence voltage is therefore established, and a common-mode voltage compensation method is combined with an improved Kalman-filter-based harmonic extraction and closed-loop suppression strategy. The switching performance of the modulation is evaluated by counting the switching transitions of the twelve IGBTs and by comparing the zero-sequence current and phase-current spectra. Experiments on a 10 kHz prototype demonstrate that the proposed method substantially reduces zero-sequence harmonics while retaining the reduced-switching characteristic of the original ASHCPWM. Full article

This study examined the effect of breathing rate (BR) on prefrontal cortical (PFC) activation during single-task and dual-task (e.g., Trail Walking Task (TWT)) conditions in a diverse aging population. We hypothesized that higher BR would be associated with increased PFC activation, particularly during dual-task walking. In this cross-sectional study, 45 adults (mean age 50.31 ± 20.19 years; 27 females) walked on an instrumented treadmill at a self-selected pace under single-task and TWT conditions. Resting BR was measured using a smart shirt, and functional near-infrared spectroscopy assessed PFC oxygenated hemoglobin (HbO₂) and deoxygenated hemoglobin (Hb) during TWTA (numbers only) and TWTB (numbers and letters). Linear mixed-effects models evaluated the effects of BR, task, and their interaction on PFC activation, controlling for age and cardiorespiratory fitness (VO₂ max). The results showed significant differences in Hb and HbO₂ between single- and dual-task conditions across BR levels. Individuals with higher BR exhibited greater PFC activation during TWT, whereas lower BR was associated with reduced activation and more efficient attentional resource allocation. BR was not associated with walking speed. These findings suggest that higher BR reflects increased neural effort during dual-task walking and highlight breathing as a potential modifiable factor influencing cognitive-motor performance and fall risk. Full article

Background/Objectives: Parkinson’s disease (PD) is a neurodegenerative disease with a heterogeneous nature. Many molecular pathways are engaged in PD pathogenesis. Granzyme B (GrB) is an enzyme released by CTLs and has roles in neuroinflammation, axonal degeneration, demyelination, and neuronal ischemic death. However, little is reported about its role in PD pathogenesis and deterioration. To evaluate the role of GrB in PD. Method: A total of 94 participants were recruited from outpatient neurology clinics (47 PD patients and 47 healthy controls). Serum GrB levels were measured using ELISA. Results: Among PD patients, the age of onset was 57.30 ± 4.92 years, and the duration of illness was 7.13 ± 4.20 years. Sociodemographic data revealed statistically significant associations between the development of PD and HCV infection (0.001*), family history of neuropsychiatric illness (0.004*), and/or PD (0.028*). MoCA TS showed a significant reduction in cognitive performance among PD patients even after correction (p < 0.001*). H_Y score showed that >50% of PD patients were clustered in stages 1 and 2, and motor scores showed a substantial reduction. GrB levels were markedly increased among PD patients compared with the control group (1721.4 ± 588.8 pg/mL vs. 418.4 ± 131.3 pg/mL) (p < 0.001*). This indicates a strong association between elevated granzyme B and PD. However, no statistically significant correlations were observed between GrB levels and the parameters studied. Conclusion: GrB levels are significantly elevated among PD patients. This reflects underlying immune activation or inflammatory processes associated with the progression of PD. GrB showed a promising diagnostic association with PD, but was not correlated with disease severity, activity, or duration in this cohort. Formal evaluation of GrB’s diagnostic accuracy in an independent, disease-control-inclusive cohort is warranted before it can be considered a diagnostic biomarker. Full article

This study proposes a motor imagery brain-computer interface (MI-BCI) experimental paradigm for right-hand steering movements in driving scenarios. The proposed paradigm aligns more closely with actual steering behavior at the action level, aiming to address the mismatch between conventional motor imagery and actual driving tasks. To validate the feasibility of this paradigm, motor imagery EEG data were collected from 12 participants using an eight-electrode low-cost OpenBCI device and systematically analyzed. Time-frequency analysis revealed that, during right-hand steering imagery, electrode C3 exhibited event-related desynchronization (ERD) in both the mu and beta bands, whereas electrode C4 also showed ERD in the beta band but exhibited event-related synchronization predominantly in the mu band. Spatial pattern analysis further indicated that common spatial pattern (CSP) yielded discriminative weight distributions over the central region associated with steering direction, suggesting that the paradigm can elicit neural pattern differences corresponding to steering direction. On this basis, baseline methods, including CSP, FBCSP, EEGNet, EEG-TCNet, and ATCNet were employed to evaluate the separability of left- and right-turn steering tasks performed with the right hand under this paradigm. To improve decoding performance, the SMB-TCSAN model was developed, integrating Sinc convolution, multi-branch temporal convolution, and a multi-head self-attention mechanism, achieving an average classification accuracy of 66.80% on the steering-direction dataset. This study preliminarily validates the feasibility of low-cost, few-channel EEG devices in steering motor imagery decoding, providing a conceptual design reference for future research on driving-assistive BCI systems. Full article

Conventional proportional integral (PI) current controllers used in field-oriented control (FOC) have an acceptable performance under linear motor models, but their performance degrades when nonlinear machine characteristics are taken into account. This paper presents a comparative evaluation of PI and twin delayed deep deterministic policy gradient (TD3)-based current control for a permanent magnet synchronous motor (PMSM) drive using a high-fidelity nonlinear motor model. The first step involves modeling and validating a PMSM in JMAG using finite element analysis, assessing its electromagnetic properties. Subsequently, JMAG-RT is utilized to generate nonlinear flux, torque, and inductance maps, which are implemented as lookup tables (LUTs) in MATLAB/Simulink (R2026a) to accurately capture the motor’s nonlinear behavior. Then, the performance of both the PI and TD3 controllers is tested with both lumped-parameter and nonlinear motor models. The results show that the PI controller has strong speed oscillation and current fluctuation under nonlinear operating conditions, but the TD3-based controller has better tracking accuracy and current regulation performance. In terms of integral absolute error (IAE), compared to PI-based nonlinear motor parameters, the TD3 control framework provides a 17.6% reduction in speed IAE, and a minor 0.4% decrease in q-axis current IAE, resulting in accurate reference tracking under nonlinear operating conditions. These findings demonstrate the effectiveness of reinforcement learning for enhancing the performance of PMSM drives when high-fidelity nonlinear motor models are considered. Full article

Background: Cerebral palsy is associated with complex motor impairments, including spasticity, reduced joint mobility, and functional limitations. Longitudinal data describing changes during sustained rehabilitation remain limited. Objective: We evaluated longitudinal changes in muscle tone, joint mobility, and gross motor function in children with spastic diplegic cerebral palsy participating in a structured 12-month rehabilitation program. Methods: This prospective, uncontrolled observational study included 94 children aged 5–10 years, selected from 116 children screened for eligibility. Assessments were performed at baseline (T0), 6 months (T1), and 12 months (T2). Outcomes included muscle tone (Modified Ashworth Scale, MAS), joint mobility (goniometry), and gross motor function (GMFM-88). Longitudinal changes were assessed using the Friedman test, followed by Bonferroni-adjusted post hoc pairwise comparisons between T0–T1, T0–T2, and T1–T2. Associations were assessed using Spearman’s rank correlation coefficient (ρ). Results: Significant longitudinal differences were observed across the evaluated outcomes (overall Friedman tests, p < 0.0001). Median MAS scores decreased from 3 at T0 to 2 at T2 across the evaluated muscle groups. Hip abduction increased, knee extension deficit decreased, and ankle dorsiflexion improved over the study period. GMFM-88 scores also increased longitudinally. Moderate-to-strong associations were observed between muscle spasticity and joint mobility parameters (ρ = −0.881 to 0.682). Conclusions: Favorable longitudinal changes in muscle tone, joint mobility, and gross motor function were observed during participation in the 12-month rehabilitation program. Given the uncontrolled observational design, these changes cannot be attributed exclusively to the rehabilitation intervention. Full article

According to the Movement Disorder Society Unified Parkinson’s Disease Rating Scale (MDS-UPDRS), Parkinson’s disease (PD) is the second most prevalent neurodegenerative disorder among older adults, surpassed only by Alzheimer’s disease. PD is characterized by a heterogeneous combination of motor and non-motor symptoms, with involuntary movements constituting a central clinical defining feature. The prodromal phase presents substantial diagnostic challenges, as early symptoms are often mild, nonspecific, and commonly misinterpreted because overt motor signs are absent. Significantly, these early alterations can precede formal diagnosis by up to two decades, highlighting the urgent need for effective early detection strategies. In this work, we propose a functional Magnetic Resonance Imaging (fMRI)-based image-processing and machine learning framework for early detection of PD. Statistical features are extracted to train a set of 25 machine learning kernels, from which the most accurate model is selected. We then apply the Minimum Redundancy Maximum Relevance (mRMR) algorithm to identify the most descriptive fMRI frames for each subject. We introduce a hierarchical classification strategy: an initial binary classification to distinguish control subjects from prodromal + PD cases, followed by a second binary classification to discriminate between Prodromal and PD. The experimental results demonstrate a maximum precision of $96.3$% with 16 axial slices and $86.4$% with a single slice, indicating the effectiveness of the preprocessing strategy and its potential as a non-invasive biomarker for early PD detection, even with reduced input data. These findings suggest that high classification performance can be achieved with a limited number of fMRI slices, facilitating data acquisition and reducing subject burden in clinical studies. Full article

The rapid development of artificial intelligence (AI) has created new opportunities for improving the performance, robustness, and efficiency of electrical machine drive systems. Among AI-based approaches, fuzzy logic control (FLC) has attracted considerable attention because of its ability to handle nonlinear dynamics, parameter uncertainties, and external disturbances without relying on an accurate mathematical model. This review systematically examines FLC-based control strategies for electrical machine drives, with particular emphasis on induction motors, switched reluctance motors, permanent-magnet synchronous motors, synchronous reluctance motors, and brushless DC motors. The review follows the PRISMA 2020 framework, and the selected studies are analyzed according to machine type, FLC architecture, control strategy, optimization method, implementation platform, and validation approach. The reviewed evidence indicates that FLC-based strategies can improve dynamic response, tracking accuracy, robustness, and torque regulation under the specific conditions reported in the literature. Hybrid approaches combining FLC with field-oriented control, direct torque control, sliding-mode control, model predictive control, neural networks, ANFIS, and optimization algorithms provide additional opportunities for adaptation and parameter tuning. However, the reported performance is strongly dependent on machine topology, controller architecture, tuning methodology, computational requirements, and validation platform. The review also identifies important limitations, including the lack of standardized benchmarking, computational complexity, dependence on expert knowledge, and limited HIL and experimental validation of several advanced approaches. Emerging directions include Type-2 and higher-order fuzzy systems, neuro-fuzzy and hybrid AI controllers, data-driven optimization, digital-twin-assisted control, edge computing, and hardware-oriented implementation. The objective of this review is to provide a structured and critical synthesis of the existing evidence, clarify the evolution and practical applicability of AI-driven FLC approaches, and identify research priorities for reliable, computationally efficient, and experimentally validated electrical machine control. Full article

Mating performance requires complex motor skills and inter-individual synchronization, yet the developmental pathways of copulatory competency remain poorly understood. Reverse mounting, where females mount males, is integrated into the behavioural toolkits of various monomorphic, long-lived birds, but its adaptive function remains debated. Using a 35-year demographic dataset of a wild griffon vulture (Gyps fulvus) population, we tested whether reverse mounting functions as a mechanism of ‘sexual tutoring’ during early reproductive life. Longitudinal observations of individuals marked as nestlings (n = 455) and tracked through their reproductive stage revealed that, out of 1498 total observed copulations for them, 1.5% corresponding to 23 cases were reverse copulations involving 9 marked individuals of known exact age. Reverse mounting showed a significant decrease with age, being concentrated in earlier stages with a mean age of 7 years, where 95.7% of these interactions featured an adult female paired with a younger male, and virtually disappearing by 12 years of age with the exception of a single record involving a 25-year-old individual. Subadult females-on-adult males mounting was never observed. While breeding attempts featuring observed reverse copulations showed reduced laying probability, final breeding success was correlated with the individual’s chronological age rather than the behaviour itself. Our findings provide correlational evidence consistent with the hypothesis that reverse mounting may act as a socially mediated learning process. In large-bodied, monomorphic species with high mate fidelity, experienced partners could mitigate asymmetric skills by tutoring youth, potentially accelerating the acquisition of necessary copulatory dexterity. Full article

Background: Our objective was to evaluate the clinical effects on walking function and symmetry of combined knee-powered exoskeleton-assisted gait training (EAGT) and conventional physiotherapy versus conventional therapy alone in patients with subacute stroke. Methods: This randomized pilot trial enrolled patients 1–3 months post-stroke with Functional Ambulation Classification scores of 3 or 4. Participants were randomly assigned in a 1:1 ratio to receive EAGT plus conventional therapy or conventional therapy alone for 4 weeks. Outcomes included gait speed (10MWT), endurance (6MWT), and motor function (FMA-LE), assessed at baseline, 4 weeks, and 16 weeks. Accelerometer-based spatiotemporal gait parameters were analyzed in the EAGT group. Results: Twenty-four participants (EAGT: 17, control: 7) completed the study. No significant group × time interaction was observed for gait speed, walking endurance, or lower-extremity motor function, indicating that changes over time did not differ significantly between the EAGT and control groups. In the EAGT group, increased gait speed was accompanied by augmented cadence (p < 0.001) without significant changes in stride length or gait symmetry. Conclusions: From this preliminary pilot study, knee-powered EAGT improved walking performance in subacute stroke but did not demonstrate superiority over conventional rehabilitation. Gains were accompanied by cadence augmentation without restoration of gait symmetry. Larger randomized trials are needed to validate these preliminary findings. Full article

The rhizosheath is a dynamic plant–soil interface in which root traits, microbial activity, and soil physical properties jointly regulate plant adaptation to drought and nutrient limitation. Despite the growing interest in this field, it remains conceptually fragmented. This study mapped the development, structure, and emerging directions of rhizosheath research at the intersection of microbiome interactions, stress adaptation, and root-trait genetics. Bibliometric and science-mapping analyses were performed on 136 publications (2015–2026) retrieved from the Web of Science Core Collection. Using the Bibliometrix framework, we examined publication dynamics, collaboration networks, citation patterns, keyword co-occurrence, and thematic structures. The dataset comprised 136 publications, 6366 cited references, and 723 authors, with 54.41% international co-authorship. Logistic modeling described the accumulation of publications through 2025, identifying a growth inflection at 2022.54; a reliable saturation level could not be estimated because the 2026 data cover only a partial year. Document coupling resolved nine clusters dominated by soil–root interface processes (n = 51; 1358 citations) and plant–microbe interactions (n = 18; 895 citations). Thematic analysis positioned soil and rhizosheath as central domains and identified water stress as a key motor theme, whereas mucilage, hydraulic functioning, and microbiome assembly emerged as recent trends. Rhizosheath research is transitioning from descriptive characterization toward more integrated, mechanistic perspectives linking root traits, soil processes, and microbial dynamics. Progress will depend on resolving genotype × soil × microbiome interactions and advancing field-based cross-scale phenotyping to support climate-resilient cropping systems. Full article

Background and Objectives: Stroke, a neurological disorder caused by damage to the central nervous system, is commonly accompanied by impairments in motor and sensory function and muscle strength. We compared the effects of progressive squat exercises performed under varying support-surface stability conditions on quadriceps femoris muscle thickness in patients 6 months to less than 1 year after stroke, using rehabilitative ultrasound imaging. Materials and Methods: This randomized, open-label, parallel-group controlled trial assessed 36 patients with stroke for eligibility. Four individuals were excluded before randomization; the remaining 32 participants were randomly assigned in a 1:1 ratio to either the unstable-support-surface squat exercise group (n = 16) or the stable-support-surface squat exercise group (n = 16). Both groups performed progressive squat exercises five times per week for 4 weeks. Quadriceps femoris muscle thickness was measured before and after the intervention using rehabilitative ultrasound imaging. Results: Between-group comparisons of changes from baseline showed significantly greater increases in rectus femoris and vastus medialis muscle thickness in the unstable support-surface squat exercise group than in the stable support-surface squat exercise group (p < 0.05). No significant between-group differences were observed in vastus intermedius or vastus lateralis muscle thickness (p > 0.05). Conclusions: Progressive squat exercise performed on an unstable support surface resulted in greater increases in rectus femoris and vastus medialis muscle thickness than the same exercise performed on a stable support surface. However, these findings are limited to muscle thickness and should not be interpreted as evidence of improved muscle strength or functional performance. Full article