Search Results (571)

With the advancement of task-oriented reinforcement learning (RL), the capability of quadruped robots for motion generation and complex task completion has significantly improved. However, current control strategies require extensive domain expertise and time-consuming design processes to acquire operational skills and achieve multi-task motion control, often failing to effectively manage complex behaviors composed of multiple coordinated actions. To address these limitations, this paper proposes a motion policy generation method for quadruped robots based on multimodal motion primitives and imitation learning. A multimodal motion library was constructed using 3D engine motion design, motion capture data retargeting, and trajectory planning. A temporal domain-based behavior planner was designed to combine these primitives and generate complex behaviors. We developed a RL-based imitation learning training framework to achieve precise trajectory tracking and rapid policy deployment, ensuring the effective application of actions/behaviors on the quadruped platform. Simulation and physical experiments conducted on the Lite3 quadruped robot validated the efficacy of the proposed approach, offering a new paradigm for the deployment and development of motion strategies for quadruped robots. Full article

Accurate braking-distance prediction for heavy-duty multi-axle trucks remains challenging due to the large gross vehicle weight, tandem-axle interactions, and strong transient load transfer during emergency braking. Recent studies on tire–road friction estimation, commercial-vehicle braking control (EBS/AEBS), and weigh-in-motion (WIM) sensing have highlighted that unmeasured vertical-load dynamics and time-varying friction are key sources of prediction uncertainty. To address these limitations, this study proposes an integrated sensing–simulation–AI framework that combines real-time axle-load estimation, full-scale robotic braking tests, fused road-friction sensing, and physics-consistent machine-learning modeling. A micro-electro-mechanical systems (MEMS)-based load-angle sensor was installed on the leaf-spring panel linking tandem axles, enabling the continuous estimation of dynamic vertical loads via a polynomial calibration model. Full-scale on-road braking tests were conducted at 40–60 km/h under systematically varied payloads (0–15.5 t) using an actuator-based braking robot to eliminate driver variability. A forward-looking optical friction module was synchronized with dynamic axle-load estimates and deceleration signals, and additional scenarios generated in a commercial ASM environment expanded the operational domain across a broader range of friction, grade, and loading conditions. A gradient-boosting regression model trained on the hybrid dataset reproduced measured stopping distances with a mean absolute error (MAE) of 1.58 m and a mean absolute percentage error (MAPE) of 2.46%, with most predictions falling within ±5 m across all test conditions. The results indicate that incorporating real-time dynamic axle-load sensing together with fused friction estimation improves braking-distance prediction compared with static-load assumptions and purely kinematic formulations. The proposed load-aware framework provides a scalable basis for advanced driver-assistance functions, autonomous emergency braking for heavy trucks, and infrastructure-integrated freight safety management. All full-scale braking tests were carried out at approximately 60% of the nominal service-brake pressure, representing non-panic but moderately severe braking conditions, and the proposed model is designed to accurately predict the resulting stopping distance under this prescribed braking regime rather than to minimize the absolute stopping distance itself. Full article

All-sky cameras provide continuous hemispherical observations essential for atmospheric monitoring and observatory operations; however, automated classification of sky conditions in tropical environments remains challenging due to strong illumination variability, atmospheric scattering, and overlapping thin-cloud structures. This study proposes EfficientNet-Attention-SPP Multi-scale Network (EASMNet), a physics-aware deep learning framework for robust all-sky scene classification using hemispherical imagery acquired at the Thai National Observatory. The proposed architecture integrates Squeeze-and-Excitation (SE) blocks for radiometric channel stabilization, the Convolutional Block Attention Module (CBAM) for spatial–semantic refinement, and Spatial Pyramid Pooling (SPP) for hemispherical multi-scale context aggregation within a fully fine-tuned EfficientNetB7 backbone, forming a domain-aware atmospheric representation framework. A large-scale dataset comprising 122,660 RGB images across 13 day–night sky-scene categories was curated, capturing diverse tropical atmospheric conditions including humidity, haze, illumination transitions, and sensor noise. Extensive experimental evaluations demonstrate that the EASMNet achieves 93% overall accuracy, outperforming representative convolutional (VGG16, ResNet50, DenseNet121) and transformer-based architectures (Swin Transformer, Vision Transformer). Ablation analyses confirm the complementary contributions of hierarchical attention and multi-scale aggregation, while class-wise evaluation yields F1-scores exceeding 0.95 for visually distinctive categories such as Day Humid, Night Clear Sky, and Night Noise. Residual errors are primarily confined to physically transitional and low-contrast atmospheric regimes. These results validate the EASMNet as a reliable, interpretable, and computationally feasible framework for real-time observatory dome automation, astronomical scheduling, and continuous atmospheric monitoring, and provide a scalable foundation for autonomous sky-observation systems deployable across diverse climatic regions. Full article

Accurate identification of road roughness is pivotal for optimizing vehicle suspension control and enhancing passenger comfort. However, existing data-driven methods often struggle to balance classification accuracy with the strict computational constraints of real-time onboard monitoring. To address this challenge, this paper proposes a lightweight and robust road roughness classification framework utilizing a single sprung mass accelerometer. First, to overcome the scarcity of labeled real-world data and the limitations of linear models, a high-fidelity co-simulation platform combining CarSim and Simulink is established. This platform generates physically consistent vibration datasets covering ISO A–F roughness levels, effectively capturing nonlinear suspension dynamics. Second, we introduce DB-MLP, a novel Dual-Branch Multi-Layer Perceptron architecture. In contrast to computationally intensive Transformer or RNN-based models, DB-MLP employs a dual-branch strategy with multi-resolution temporal projection to efficiently capture multi-scale dependencies, and integrates dual-domain (time and position-wise) feature transformation blocks for robust feature extraction. Experimental results demonstrate that DB-MLP achieves a superior accuracy of 98.5% with only 0.58 million parameters. Compared to leading baselines such as TimeMixer and InceptionTime, our model reduces inference latency by approximately 20 times (0.007 ms/sample) while maintaining competitive performance on the specific road classification task. This study provides a cost-effective, high-precision solution suitable for real-time deployment on embedded vehicle systems. Full article

Smart city infrastructures are evolving from centralized cloud systems to distributed Cyber-Physical Systems of Systems (CPSoS), requiring integration across heterogeneous administrative domains. This work presents a flexible, modular, multi-domain architecture for automated orchestration and management of IoT services across heterogeneous environments. It relies on a recursive federation model, where autonomous local domains manage their own resources while higher-level components coordinate cross-domain operations. Interoperability is achieved through standardized interfaces using TM Forum Open APIs and ETSI NGSI-LD, while a Secure Integration Fabric enables secure, policy-based coordination across public and private domains. The architecture is validated in a real-world Smart Waste Management pilot, demonstrating support for flexible workflows, cross-platform collaboration, real-time decision-making, and avoidance of vendor lock-in. Experimental results show that dynamic, context-driven service orchestration improves scalability, interoperability, and resource efficiency compared to static deployments. Full article

Floating-base robotic systems rely critically on their inertial geometry to maintain stability and regulate interaction forces in the absence of fixed ground constraints. Their control authority additionally depends on the placement and orientation of actuators relative to the center of mass, which determines the moment arms through which thrust or force inputs generate stabilizing actions. This paper develops a general theoretical framework showing that internal mass shifting provides a powerful, domain-independent mechanism for reshaping global system dynamics. Through geometric principles governing center-of-mass placement, moment-arm modification, and inertia redistribution, mass shifting enhances passive stability, reduces the torque induced by external disturbances, and improves the controllability of interaction-intensive tasks. The theory is first examined in a buoyancy-driven simulation of a two-mass floating body subjected to multi-sine wave excitation, which isolates the hydrostatic effects of center-of-mass displacement. To validate the generality of these principles, we further demonstrate their applicability in a radically different domain through real-world experiments on the AeroBull aerial robot, a multirotor platform equipped with an internal mass-shifting mechanism for aerial manipulation. Across both aquatic and aerial settings, mass shifting consistently improves stability, reduces control effort, and increases achievable interaction forces. These results establish internal mass redistribution as a platform-agnostic strategy for enhancing the stability and resilience of floating-base robots operating in uncertain and physically demanding environments. Full article

Digital signal processing plays a central role in intelligent railway communications under high-mobility, strong-multipath, and time-varying-channel conditions. This review surveys representative techniques for multi-carrier modulation, precoding, index modulation, and chaos-inspired physical layer security and highlights their mathematical foundations. Core themes include transform-domain representations typified by time–frequency analysis, linear-algebraic formulations of precoding and equalization, combinatorial structures underlying index mapping and spectral efficiency gains, and nonlinear dynamical systems theory of chaotic encryption. The methods are compared in terms of bit error performance, peak-to-average power ratio, spectral efficiency, computational complexity, and information security, with emphasis on railway-specific deployment constraints. The synergistic application of these methods with intelligent railway transportation systems is expected to enhance the overall performance of railway transportation systems in terms of transmission efficiency, reliability, and security. It provides critical technological support for the efficient and secure operation of next-generation intelligent transportation systems. Full article

Remote sensing plays a vital role in monitoring environmental change in Antarctica, offering non-invasive insights into ice dynamics, biodiversity, and fragile ecosystems. Harsh conditions, limited field access, and logistical challenges result in sparse, noisy, and often unlabelled datasets, posing major obstacles for machine learning (ML) approaches. Data scarcity remains a fundamental challenge for uncrewed aerial vehicle (UAV)-based ecological monitoring. While ML models in other Earth observation domains demonstrate state-of-the-art performance, their applicability in Antarctic and polar regions’ settings is limited. This paper reviews the intersection of ML and UAV-based remote sensing in Antarctica under extreme data constraints. We surveyed recent strategies designed to overcome these limitations, including self-supervised learning, physics-informed modelling, and foundation models. Results highlight a notable gap, as polar environments remain excluded from global datasets and benchmarks due to the extensive data requirements of large-scale models. Opportunities exist where multimodal and multi-scale generalisation can enhance cross-domain adaption to data-scarce use cases. Unlike prior reviews on general remote sensing or task-specific polar studies, this work uniquely underscores the need for Antarctic representation in global ML advances, positioning Antarctica as a frontier testbed for machine learning in extreme, inaccessible, and under-resourced fields. Full article

Objectives: Field-based testing is commonly used to evaluate key physical qualities related to soccer performance. However, limited research has examined the degree of shared variance among measures of aerobic capacity, change of direction (COD), and explosive power in youth athletes. This study investigated the relationships between the 20 m shuttle run (20MSR), T-test (TT), and vertical countermovement jump (CMJ) to determine their unique and overlapping contributions to each other’s performance in competitive youth soccer players. Methods: Twenty-five competitive male youth soccer players (13.7 ± 0.8 years) completed standardized assessments of TT, CMJ, and 20MSR during pre-season evaluations. Pearson correlations and hierarchical multiple regression analyses were used to examine associations and independent variance explained among the performance measures. Results: Large, significant correlations were observed between TT and CMJ (r = −0.65, p < 0.001), TT and 20MSR (r = −0.59, p < 0.001), and CMJ and 20MSR (r = 0.53, p = 0.007). CMJ explained 42.3% of TT variance, whereas adding 20MSR did not significantly improve model fit (ΔR² = 0.087, p = 0.062). Across models, aerobic capacity did not contribute significant unique variance beyond neuromuscular performance. Conclusions: COD and lower-body power share a common physiological foundation in youth soccer athletes, while aerobic capacity represents a distinct performance domain. When field tests are administered under applied conditions typical of youth soccer environments, TT and CMJ demonstrate substantial shared variance, whereas 20MSR remains largely independent. Therefore, the findings support the continued use of multi-modal testing batteries in practice. Full article

Carbon capture and storage (CCS) is essential for achieving net-zero emissions, yet amine-based capture systems face significant challenges including high energy penalties (20–30% of power plant output) and operational costs ($50–120/tonne CO${}_2$). This study develops and validates a novel multi-scale Digital Twin (DT) framework integrating Physics-Informed Neural Networks (PINNs) to address these challenges through real-time optimization. The framework combines molecular dynamics, process simulation, computational fluid dynamics, and deep learning to enable real-time predictive control. A key innovation is the sequential training algorithm with domain decomposition, specifically designed to handle the nonlinear transport equations governing CO${}_2$ absorption with enhanced convergence properties.The algorithm achieves prediction errors below 1% for key process variables (R${}^2$> 0.98) when validated against CFD simulations across 500 test cases. Experimental validation against pilot-scale absorber data (12 m packing, 30 wt% MEA) confirms good agreement with measured profiles, including temperature (RMSE = 1.2 K), CO${}_2$ loading (RMSE = 0.015 mol/mol), and capture efficiency (RMSE = 0.6%). The trained surrogate enables computational speedups of up to four orders of magnitude, supporting real-time inference with response times below 100 ms suitable for closed-loop control. Under the conditions studied, the framework demonstrates reboiler duty reductions of 18.5% and operational cost reductions of approximately 31%. Sensitivity analysis identifies liquid-to-gas ratio and MEA concentration as the most influential parameters, with mechanistic explanations linking these to mass transfer enhancement and reaction kinetics. Techno-economic assessment indicates favorable investment metrics, though results depend on site-specific factors. The framework architecture is designed for extensibility to alternative solvent systems, with future work planned for industrial-scale validation and uncertainty quantification through Bayesian approaches. Full article

Waste management remains a major challenge worldwide, as rapidly expanding urban populations put greater pressure on traditional disposal methods such as landfilling and incineration. Plasma-based waste treatment offers an innovative, sustainable waste-to-energy solution capable of converting a wide range of waste types. Although plasma technologies provide significant environmental benefits, such as greatly reducing waste volume and emissions compared to conventional approaches, their widespread adoption faces notable economic hurdles. Primary among these is high operational cost due to system inefficiencies. These costs mainly arise from energy losses within the plasma torch, energy consumed during plasma torch tuning with the plasma reactor, and power inefficiencies when processing unsuitable waste loads. These issues not only increase costs but also impact process stability, which can influence stakeholder support and the technology’s commercial potential. Optimizing the process through simulation presents an effective approach to overcoming this inefficiency. However, relying solely on these advanced tools can be time-consuming and requires substantial domain expertise, creating a bottleneck in design and optimization. This paper introduces a new integrated platform combining COMSOL Multiphysics v6.2, Ansys Fluent 2024 R1, and Aspen Plus v12.1 to address these challenges. Using a genetic algorithm, the platform automates the complex task of designing an optimal plasma torch, optimizes it for peak performance, and dynamically adjusts plasma conditions. This intelligent optimization system aims to maximize energy output and process efficiency, directly tackling key cost-related issues. Full article

With the development of aerospace technology, hypersonic flight vehicles are evolving towards larger size, lighter weight, and higher performance. Their cross-domain maneuverability and extreme flight environment led to the rigid–flexible coupling effect and became the core bottleneck restricting performance improvement, seriously affecting flight stability and control accuracy. This paper systematically reviews the research status in the field of control for high-speed rigid–flexible coupling aircraft and conducts a review focusing on two core aspects: dynamic modeling and control strategies. In terms of modeling, the modeling framework based on the average shafting, the nondeformed aircraft fixed-coordinate system, and the transient coordinate system is summarized. In addition, the dedicated modeling methods for key issues, such as elastic mode coupling and liquid sloshing in the fuel tank, are also presented. The research progress and challenges of multi-physical field (thermal–structure–control, fluid–structure–control) coupling modeling are analyzed. In terms of control strategies, the development and application of linear control, nonlinear control (robust control, sliding mode variable structure control), and intelligent control (model predictive control, neural network control, prescribed performance control) are elaborated. Meanwhile, it is pointed out that the current research has limitations, such as insufficient characterization of multi-physical field coupling, neglect of the closed-loop coupling characteristics of elastic vibration, and lack of adaptability to special working conditions. Finally, the relevant research directions are prospected according to the priority of “near-term engineering requirements–long-term frontier exploration”, providing Refs. for the breakthrough of the rigid–flexible coupling control technology of the new-generation high-speed aircraft. Full article

The expansion of renewables in modern power systems and the coordinated development of upstream and downstream industrial chains are promoting a shift on the utility side from traditional settlement by energy toward operation driven by data and models. Industrial electricity consumption data exhibit pronounced multi-scale temporal structures and sectoral heterogeneity, which makes unified long-term load and generation forecasting while maintaining accuracy, interpretability, and scalability a challenge. From a modern system identification perspective, this paper proposes a System Identification in Adaptive Bayesian Framework (SIABF) for medium- and long-term industrial load forecasting based on daily freeze electricity time series. By combining daily aggregation of high-frequency data, frequency domain analysis, sparse identification, and long-term extrapolation, we first construct daily freeze series from 15 min measurements, and then we apply discrete Fourier transforms and a spectral complexity index to extract dominant periodic components and build an interpretable sinusoidal basis library. A sparse regression formulation with ${\ell }_1$ regularization is employed to select a compact set of key basis functions, yielding concise representations of sector and enterprise load profiles and naturally supporting multivariate and joint multi-sector modeling. Building on this structure, we implement a state-space-implicit physics-informed Bayesian forecasting model and evaluate it on real data from three representative sectors, namely, steel, photovoltaics, and chemical, using one year of 15 min measurements. Under a one-month-ahead evaluation using one year of 15 min measurements, the proposed framework attains a Mean Absolute Percentage Error (MAPE) of 4.5% for a representative PV-related customer case and achieves low single-digit MAPE for high-inertia sectors, often outperforming classical statistical models, sparse learning baselines, and deep learning architectures. These results should be interpreted as indicative given the limited time span and sample size, and broader multi-year, population-level validation is warranted. Full article

This review synthesizes the state of the art in flapping foil technology and bridges the distinct engineering domains of bio-inspired propulsion and power generation via flow energy harvesting. This review is motivated by the observation that propulsion and power-generation studies are frequently presented separately, even though they share common unsteady vortex dynamics. Accordingly, we adopt a unified unsteady-aerodynamic perspective to relate propulsion and energy-extraction regimes within a common framework and to clarify their operational duality. Within this unified framework, the feathering parameter provides a theoretical delimiter between momentum transfer and kinetic energy extraction. A critical analysis of experimental foundations demonstrates that while passive structural flexibility enhances propulsive thrust via favorable wake interactions, synchronization mismatches between deformation and peak hydrodynamic loading constrain its benefits in power generation. This review extends the analysis to complex and non-homogeneous environments and identifies that density stratification fundamentally alters the hydrodynamic performance. Specifically, resonant interactions with the natural Brunt–Väisälä frequency of the fluid shift the optimal kinematic regimes. The present study also surveys computational methodologies and highlights a paradigm shift from traditional parametric sweeps to high-fidelity three-dimensional (3D) Large-Eddy Simulations (LESs) and Deep Reinforcement Learning (DRL) to resolve finite-span vortex interconnectivities. Finally, this review outlines the critical pathways for future research. To bridge the gap between computational idealization and physical reality, the findings suggest that future systems prioritize tunable stiffness mechanisms, multi-phase environmental modeling, and artificial intelligence (AI)-driven digital twin frameworks for real-time adaptation. Full article

Modern power distribution networks are increasingly challenged with nonlinear operating conditions, the high penetration of distributed energy resources, and conflicting operational objectives such as loss minimization and voltage regulation. Existing load-flow optimization approaches often suffer from slow convergence, premature stagnation in non-convex search spaces, and limited robustness when handling conflicting multi-objective performance criteria under fixed network constraints. To address these challenges, this paper proposes a Fractional Multi-Objective Load Flow Optimizer (FMOLFO), which integrates a fractional-order numerical regularization mechanism with an adaptive Pareto-based Differential Evolution framework. The fractional-order formulation employed in FMOLFO operates over an auxiliary iteration domain and serves as a numerical regularization strategy to improve the sensitivity conditioning and convergence stability of the load-flow solution, rather than modeling the physical time dynamics or memory effects of the power system. The optimization framework simultaneously minimizes physically consistent active power loss and voltage deviation within existing network operating constraints. Extensive simulations on IEEE 33-bus and 69-bus benchmark distribution systems demonstrate that FMOLFO achieves an up to 27% reduction in active power loss, improved voltage profile uniformity, and faster convergence compared with classical Newton–Raphson and metaheuristic baselines evaluated under identical conditions. The proposed framework is intended as a numerically enhanced, optimization-driven load-flow analysis tool, rather than a control- or dispatch-oriented optimal power flow formulation. Full article