Search Results (68)

Wrist rehabilitation requires high precision, haptic transparency, and accurate alignment with the human joint’s physiological center of rotation. Conventional robotic systems often suffer from high moving inertia or joint misalignment. This paper presents the design and kinematic validation of a novel 3-DOF spherical parallel exoskeleton featuring base-fixed actuators. By mounting all actuators to a fixed base, the proposed architecture significantly reduces moving mass, achieving a low-inertia response critical for safe patient–robot interaction. The “virtual center” concept eliminates physical central joints, enabling a compact design completed by the user’s anatomy. To perform the kinematic and singularity analyses of the manipulator, two distinct models were used: Rotated Frame Based (RFB) and Initial Frame Based (IFB). Performance metrics, namely the manipulability index and condition number, are evaluated to assess dexterity and isotropy. Comparative kinematic analysis of Rotated (RFB) and Initial Frame Based (IFB) models confirm ideal central isotropy ($\kappa =1.0$). While RFB yields 94.14% high-dexterity ($\kappa <5.0$) and 99.78% usable ($\kappa <10.0$) workspace coverage, IFB achieves 78.80% high-dexterity ($\kappa <5.0$) and 92.18% usable ($\kappa <10.0$) workspace coverage. Analytical manipulability metrics validate singularity-free motion throughout the anatomical range. Additionally, the use of exponential rotational matrices in this paper provides systematic derivations of equations in compact form. Full article

This study considers a 1-m-1 hybrid flow shop scheduling problem that simultaneously incorporates four practical constraints: lot streaming, no-wait, blocking, and sequence-dependent setup times. Although each of these characteristics has been studied individually in the literature, their joint consideration in a single HFS model has received little attention. The problem is motivated by a real-world order sequencing problem in insulation board manufacturing, where all four constraints arise simultaneously from the production process. To formally characterize the problem, we develop a mixed-integer programming formulation that captures all operational constraints. For practical-scale problems, we propose several dispatching heuristics that can obtain sufficiently good solutions within a short computation time. We further develop a genetic algorithm as an independent solution approach to obtain high-quality solutions close to the optimum within a reasonable computation time. Computational experiments on instances generated based on real insulation board production characteristics demonstrate that the proposed algorithms outperform a benchmark greedy rule, and sensitivity analyses reveal the effects of setup time magnitude and the number of parallel machines on scheduling performance. Full article

IoT networks now handle traffic from billions of devices, and edge nodes are under constant pressure to classify that traffic and dispatch tasks within tight latency deadlines. Most existing systems treat classification and scheduling as two separate steps that run one after the other. This sequence adds unnecessary delay and breaks the feedback between the two tasks: the scheduler never sees the traffic type, and the classifier never sees the queue state. We propose ParallelEdge-AI, a system built around a shared flow encoder that feeds two task-specific heads in parallel, one for multi-class traffic classification and one for task-urgency scoring. Both heads are trained end-to-end using a joint loss that combines cross-entropy and pairwise ranking. A load-balance controller then reads the urgency scores alongside live queue lengths to decide, every 200 ms, whether a task stays local or moves to a less-loaded edge node. No global synchronisation is needed. We test the system on three real IoT datasets: RT-IoT2022, N-BaIoT, and CICIoT2023. ParallelEdge-AI reaches 97.63% accuracy and an F1-score of 97.34%, which is 3.16 percentage points above the best baseline. Inference latency is 19.62 ms per batch, the deadline-miss rate is 2.34%, and the load-imbalance index is 0.083, all three are the best results in our comparison. These numbers show that running classification and scheduling together on a shared representation is both faster and more accurate than treating them as separate problems. Full article

This paper presents the design and implementation of a field-programmable gate array (FPGA)-based System-on-Programmable-Chip (SoPC) architecture for sensorless speed control of permanent magnet synchronous motor (PMSM) drives. To enable real-time execution of the computationally intensive estimation stage, a parallelized Unscented Kalman Filter (UKF) is proposed for the joint estimation of rotor speed, position, and load torque. Unlike traditional sequential processor-based UKF implementations, the proposed parallel architecture simplifies the iterative process and significantly reduces computational latency and hardware resource utilization while preserving high estimation fidelity. This transformation reduces the number of sequential dependency stages within one estimation cycle and enables simultaneous execution of matrix operations using dedicated FPGA resources, thereby decreasing effective iteration latency. The complete control system comprises current regulators, a coordinate transformation module, a proportional–integral (PI) speed controller, and auxiliary functional blocks—all fully integrated within a single SoPC. The UKF estimator and control components are described using a hardware description language (HDL), enabling efficient hardware-level parallelism and real-time execution. The proposed system is validated through co-simulation and experimental verification on a Xilinx ZCU102 platform driving an inverter-fed PMSM. The results confirm correct real-time operation of the proposed architecture and demonstrate its feasibility for FPGA-based sensorless motor drive implementation. A detailed quantitative comparison with a fully sequential FPGA-based UKF implementation is identified as future work to further substantiate the reported latency reduction. Full article

To address the detection challenges in X-ray weld seam images caused by weak contrast, slender structures, and multi-scale coexistence, we propose MGDR-YOLO, an industrially deployable detector with four coordinated designs. First, a MultiBackbone parallel heterogeneous backbone is designed to perform complementary direction–detail modeling and lightweight context modeling under a shared shallow stem, enhancing the joint representation of fine-grained features and global semantics. Second, Gated Attention Fusion Block (GAFB) is introduced to perform selective in-scale fusion via channel gating and local–global attention mechanisms, thereby suppressing channel redundancy and noise leakage induced by naive concatenation. Third, Directional Feature Convolution (DFConv) decouples standard 2D convolution into horizontal and vertical branches and fuses them using depthwise separable convolution, substantially reducing computational cost while preserving geometric alignment. Finally, Rep Shared Convolutional Detection Head (RSCD) improves detection head consistency and inference throughput through cross-scale shared convolutions and a training-to-deployment re-parameterization scheme. The experimental results show that MGDR-YOLO significantly outperforms YOLOv11n, increasing the mean average precision (mAP) from 92.9% to 95.2%. The performance gain is most pronounced for the LP class (slender and low-contrast defects), with an mAP improvement of 10.1 percentage points. Meanwhile, the proposed model achieves a 39.4% increase in frames per second (FPS) while reducing the number of parameters by 46.2%, demonstrating superior efficiency. These results indicate that MGDR-YOLO consistently improves the accuracy and robustness of X-ray weld defect detection while maintaining real-time performance, making it well suited for resource-constrained industrial online inspection scenarios. Full article

Integrating depth sensors into mobile robots enables automated gait monitoring with potential applications in neurological disorders. This pilot study aims to evaluate the preliminary feasibility of robot-assisted gait assessment using Azure Kinect against Vicon, including individuals with multiple sclerosis, while simultaneously examining between-system, within-system, and environmental effects. A total of 20 participants were recruited to complete the eight-meter straight-line and 32 m corridor walking tests in the laboratory on the same day. Following independent data acquisition by both systems, temporal alignment was achieved through foot-event anchoring and interval trimming. On a unified timeline, 8 joint kinematic signals and 26 descriptors were extracted. Generalized estimating equations were applied, with a Bonferroni correction implemented for the 26 parallel tests to control the family error rate. The results showed: The spatiotemporal gait metrics exhibited general stability between systems and environments. Vicon better revealed variations in hip and pelvic amplitudes and restricted extension phenotypes, while the robotic system demonstrated greater sensitivity to knee posture and relative swing amplitude. The corridor environment induced an increase in stride length and a reduced step time compared to the laboratory, accompanied by a greater peak of hip and knee flexion and a greater forward lean of the trunk, with a largely preserved temporal organization. Within the Vicon-referenced framework, Azure Kinect-based robotic assessment demonstrated preliminary feasibility for capturing gait-related characteristics in individuals with multiple sclerosis. However, due to the limited number of analyzed MS participants, these findings should be interpreted as exploratory rather than as definitive clinical validation. The two systems exhibit complementary kinematic advantages. We recommend adopting an evaluation protocol that combines laboratory baseline with corridor validation, supplemented by descriptor-level mapping for cross-system data integration when necessary. This approach may support future tiered assessment, disease progression monitoring, and efficacy evaluation, but larger clinical cohorts are required to confirm its applicability in individuals with multiple sclerosis. Full article

In prestack seismic inversion for structurally complex areas, elastic parameters commonly show strong directional heterogeneity (layer-parallel continuity versus cross-layer discontinuities), so conventional structure-guided schemes based on isotropic regularization often struggle to achieve both numerical stability and sharp interface resolution. To address this issue, we develop a structure-oriented, direction-decomposed Lp-L2 regularization method for prestack multi-trace joint inversion of P-wave velocity (Vp), S-wave velocity (Vs), and density (ρ). Dip information extracted from poststack seismic data is used to construct a dip-guided directional operator that locally projects the Cartesian model-gradient field onto the tangential and normal directions of the structural field, corresponding to along-layer and cross-layer components, respectively. Different priors are then imposed: L2 smoothing along layers enhances lateral continuity and stabilizes the inversion, whereas a nonconvex Lp sparsity constraint across layers concentrates updates at a limited number of geological discontinuities and preserves sharp contrasts at faults and bed boundaries, thereby mitigating the over-smoothing typical of dip-guided L2 inversion. The resulting formulation is embedded in a linearized prestack Amplitude Versus Offset (AVO) framework and solved efficiently using the Alternating Direction Method of Multipliers (ADMM) algorithm. Synthetic tests and field-data applications demonstrate improved delineation of faults and thin-bed boundaries under noise, with reduced errors and higher correlation relative to a classical structure-guided L2 approach. These results indicate that the proposed method provides a practical and effective route for high-resolution prestack elastic-parameter characterization in complex tectonic settings. Full article

Edge computing has emerged as a promising paradigm to minimize latency and energy consumption while improving computational efficiency for mobile devices. Latency-sensitive applications such as autonomous driving, augmented reality, and industrial automation require ultra-low response times, making efficient task offloading a necessity in edge computing. However, distributing optimally computational tasks among edge servers remains a challenge, especially when considering latency, energy consumption, and workload balancing simultaneously. Although existing approaches have focused on one or two of these objectives, they do not provide a holistic solution that incorporates all three factors. In addition, some existing solutions do not take advantage of parallelism at the edge layer, resulting in bottlenecks and inefficient resource usage. In this paper, we propose a novel learning-based task offloading model that integrates parallel processing at the edge layer, adaptive workload balancing, and joint latency–energy optimization. Moreover, by dynamically adjusting the number of selected edge servers for parallel execution, our approach achieves optimal trade-offs between performance and resource efficiency. Our experimental setup includes several edge servers and several randomly deployed devices. It employs Apache HTTP Benchmark (AB) to generate realistic Mobile Edge Computing workloads. The obtained results show that our method outperforms existing approaches by reducing latency, lowering energy consumption, and maintaining a balanced workload across edge nodes. Full article

Rapid urban growth in many Global South cities strains waste systems and slows the shift to circular economy (CE) practice. Colombo, Sri Lanka, exemplifies this challenge, where overstretched state-led services coexist with neighborhood groups, NGOs, and informal collectors driving circular activities. This study adopts a layered social network diagnostic framework to examine how community-based waste management networks operate and how they might be reshaped to enable a city-wide CE. Using survey and interview data from 185 actors, information-sharing, collaboration, and resource-exchange networks are analyzed separately and in combination. The results reveal three principal findings: (i) Social-capital forms operate largely in parallel, with limited conversion between information, collaboration, and material exchange; (ii) the network exhibits “thin bridges and thick clusters,” in which a small number of NGO hubs mediate most cross-cluster connectivity; (iii) layers operate with mismatched coordination logics, producing gaps between awareness, collective action, and resource mobilization. As a result, ideas circulate widely but rarely translate into joint projects, local teams coordinate effectively yet remain isolated, and material flows depend on a narrow and fragile logistics spine. By diagnosing these structural misalignments, this study demonstrates a key novelty: scalable circular economy adoption depends not only on technology and policy but also on the design and alignment of underlying coordination networks. Full article

BU–MIT whistler wave injection experiments, which were conducted at Arecibo Observatory, started with the joint US–USSR Active Space Plasma Program Experiment on 24 December 1989. In this experiment, a satellite-borne VLF transmitter injected radio waves at the frequency and power of 10 kHz and 10 kW. A series of controlled whistler wave experiments with the Arecibo HF heater were subsequently carried out during 1990–1998 until the HF heater was damaged by Hurricane Georges in 1998. In these ionospheric HF heating experiments, 28.5 kHz whistler waves were launched from the nearby naval transmitter (code-named NAU) located at Aguadilla, Puerto Rico. HF heater waves were used to create ionospheric ducts (in the form of parallel-plate waveguides) to facilitate the entry of NAU whistler waves from the neutral atmosphere into the ionosphere. Conjugate whistler wave propagation experiments were conducted between Arecibo, Puerto Rico and Trelew, Argentina in 1997. After 1999, whistler wave experiments in the absence of an HF heater had been conducted. Naturally-occurring large-scale ionospheric irregularities due to spread F or Traveling Ionospheric Disturbances (TIDs) were relied on to guide NAU launched 40.75 kHz whistler waves to propagate from the ionosphere further into the radiation belts, to cause 390 keV charged-particle precipitation. A train of TIDs, resulting from the 9.2 Mw earthquake off the west coast of Sumatra, Indonesia, was observed in our 26 December 2004 Arecibo experiments, about a day after the earthquake-launched tsunami waves traveled across the Indian Ocean, then into remote parts of the Atlantic Ocean. The author’s recent research efforts, motivated by Arecibo experiments, focus on Solar Powered Microwave Transmitting Systems, to simulate Solar Energy Harvesting via Solar Power Satellite (SPS) (also known as Space Based Solar Power (SBSP)) These experiments involved a large number of the author’s BU and MIT students working on theses and participating in the Undergraduate Research Opportunities Program (UROP), in collaboration with other colleagues at several universities and national laboratories. Full article

Maritime remote sensing ship detection has long been plagued by two major issues: the failure of geometric priors due to the extreme length-to-width ratio of ships; and the sharp drop in edge signal-to-noise ratio caused by the overlapping chromaticity domain between ships and seawater, which leads to unsatisfactory accuracy of existing detectors in such scenarios. Therefore, this paper proposes an optical remote sensing ship detection model combining channel shuffling and bilinear interpolation, named CSBI-YOLO. The core innovations include three aspects: First, a group shuffling feature enhancement module is designed, embedding parallel group bottlenecks and channel shuffling mechanisms into the interface between the YOLOv8 backbone and neck to achieve multi-scale semantic information coupling with a small number of parameters. Second, an edge-gated upsampling unit is constructed, using separable Sobel magnitude as structural prior and a learnable gating mechanism to suppress low-contrast noise on the sea surface. Third, an R-IoU-Focal loss function is proposed, introducing logarithmic curvature penalty and adaptive weights to achieve joint optimization in three dimensions: location, shape, and scale. Dual validation was conducted on the self-built SlewSea-RS dataset and the public DOTA-ship dataset. The results show that on the SlewSea-RS dataset, the mAP₅₀ and mAP_50–95 values of the CSBI-YOLO model increased by 6% and 5.4%, respectively. On the DOTA-ship dataset, comparisons with various models demonstrate that the proposed model outperforms others, proving the excellent performance of the CSBI-YOLO model in detecting maritime ship targets. Full article

To investigate the strain patterns and shear failure mechanisms of concrete-filled steel tube (CFT) column-end-plate joints during construction, this study designed and fabricated a scaled model of a standard subway station section based on similarity theory. Model tests were conducted under static earth pressure loads, and a finite element model was constructed for comparative analysis. The study focused on the ultimate bearing capacity, failure modes, and force transmission mechanisms of the joint, and optimized the analysis using reinforcement parameters. The results showed significant stress concentrations at the joint corners and core, with shear forces gradually forming annular cracks in the top slab concrete. The bearing system formed by the steel tube column and flange was the primary force transmission mechanism, and even after damage, the steel tube could still effectively restrain the core concrete. Increasing the steel bar diameter and increasing the number of slab reinforcements parallel to the beam significantly improved the joint’s mechanical properties. Improving the material properties of the reinforcement increased the joint’s stress and displacement capacity. This research provides a reference for optimizing the design and construction of slab-column joints in subway stations. Full article

Facial expression recognition (FER) is a critical research direction in artificial intelligence, which is widely used in intelligent interaction, medical diagnosis, security monitoring, and other domains. These applications highlight its considerable practical value and social significance. Face expression recognition models often need to run efficiently on mobile devices or edge devices, so the research on lightweight face expression recognition is particularly important. However, feature extraction and classification methods of lightweight convolutional neural network expression recognition algorithms mostly used at present are not specifically and fully optimized for the characteristics of facial expression images, yet fail to make full use of the feature information in face expression images. To address the lack of facial expression recognition models that are both lightweight and effectively optimized for expression-specific feature extraction, this study proposes a novel network design tailored to the characteristics of facial expressions. In this paper, we refer to the backbone architecture of MobileNet V2 network, and redesign LightExNet, a lightweight convolutional neural network based on the fusion of deep and shallow layers, attention mechanism, and joint loss function, according to the characteristics of the facial expression features. In the network architecture of LightExNet, firstly, deep and shallow features are fused in order to fully extract the shallow features in the original image, reduce the loss of information, alleviate the problem of gradient disappearance when the number of convolutional layers increases, and achieve the effect of multi-scale feature fusion. The MobileNet V2 architecture has also been streamlined to seamlessly integrate deep and shallow networks. Secondly, by combining the own characteristics of face expression features, a new channel and spatial attention mechanism is proposed to obtain the feature information of different expression regions as much as possible for encoding. Thus improve the accuracy of expression recognition effectively. Finally, the improved center loss function is superimposed to further improve the accuracy of face expression classification results, and corresponding measures are taken to significantly reduce the computational volume of the joint loss function. In this paper, LightExNet is tested on the three mainstream face expression datasets: Fer2013, CK+ and RAF-DB, respectively, and the experimental results show that LightExNet has 3.27 M Parameters and 298.27 M Flops, and the accuracy on the three datasets is 69.17%, 97.37%, and 85.97%, respectively. The comprehensive performance of LightExNet is better than the current mainstream lightweight expression recognition algorithms such as MobileNet V2, IE-DBN, Self-Cure Net, Improved MobileViT, MFN, Ada-CM, Parallel CNN(Convolutional Neural Network), etc. Experimental results confirm that LightExNet effectively improves recognition accuracy and computational efficiency while reducing energy consumption and enhancing deployment flexibility. These advantages underscore its strong potential for real-world applications in lightweight facial expression recognition. Full article

In the noisy medium-scale quantum era, quantum computers are constrained by a limited number of qubits, restricted physical topological structures, and interference from environmental noise, making efficient and stable circuit scheduling a significant challenge. To improve the feasibility of quantum computing, it is essential to optimize the scheduling of quantum gates and the insertion of SWAP gates, reducing running time and enhancing computational efficiency. We propose a collaborative optimization framework that integrates the Quantum Exchange Lock Parallel Scheduler (QELPS) with the Full-level Joint Optimization SWAP Algorithm (FJOSA). In QELPS, SWAP conflict characteristics are used to adjust the layout of quantum gates across different levels while considering physical constraints and dynamically adapting to the circuit’s execution state. Quantum lock parallel technology enables the selective postponement of certain quantum gates, minimizing circuit depth and mitigating inefficiencies caused by excessive SWAP gate insertions. Meanwhile, FJOSA employs a cross-layer optimization strategy that combines heuristic algorithms with cost functions to improve gate scheduling at a global level. This approach effectively reduces quantum gate conflicts found in traditional methods and optimizes execution order, leading to better computational efficiency and circuit performance. Experimental results show that, compared to the traditional 2QAN algorithm, QELPS and FJOSA reduce additional gate insertions by 85.59% and 89.38%, respectively, while decreasing running time by 56.32% and 66.47%. These improvements confirm that the proposed method significantly enhances circuit scheduling efficiency and reduces resource consumption, making it a promising approach for optimizing quantum computation. Full article

This paper presents a joint optimization framework for multi-period planning in a Manufacturing–Remanufacturing–Transport Supply Chain (MRTSC), focusing on carbon emission reduction and economic efficiency. A novel Mixed Integer Linear Programming (MILP) model is developed to coordinate procurement, production, remanufacturing, transportation, and returns under environmental constraints, aligned with carbon tax policies and the Paris Agreement. To address uncertainty in future demand and the number of returned used products (NRUP), a two-stage approach combining forecasting and optimization is applied. Among several predictive methods evaluated, a hybrid SARIMA/VAR model is selected for its accuracy. The MILP model, implemented in CPLEX, generates optimal decisions based on these forecasts. A case study demonstrates notable improvements in cost efficiency and emission reduction over traditional approaches. The results show that the proposed model consistently maintained strong service levels through flexible planning and responsive transport scheduling, minimizing both unmet demand and inventory excesses throughout the planning horizon. Additionally, the findings indicate that carbon taxation caused a sharp drop in profit with only limited emission reductions, highlighting the need for parallel support for cleaner technologies and more integrated sustainability strategies. The analysis further reveals a clear trade-off between emission reduction and operational performance, as stricter carbon limits lead to lower profitability and service levels despite environmental gains. Full article