Search Results (704)

Underactuated systems, such as rotary and double inverted pendulums, challenge traditional control due to nonlinear dynamics and limited actuation. Classical methods like state-feedback and Linear Quadratic Regulators (LQRs) are commonly used but often require high gains, leading to excessive control effort, poor energy efficiency, and reduced robustness. This article proposes a generalized state-feedback controller with its own internal dynamics, offering greater design flexibility. To automate tuning and avoid manual calibration, we apply Bayesian Optimization (BO), a data-efficient strategy for optimizing closed-loop performance. The proposed method is evaluated on two benchmark underactuated systems, including one in simulation and one in a physical setup. Compared with standard LQR designs, the BO-tuned state-feedback controller achieves a reduction of approximately 20% in control signal amplitude while maintaining comparable settling times. These results highlight the advantages of combining model-based control with automatic hyperparameter optimization, achieving efficient regulation of underactuated systems without increasing design complexity. Full article

This study aimed to evaluate short-term clinical outcomes in COVID-19 pneumonia patients using parameters derived from a commercial deep learning-based automatic detection algorithm (DLAD) applied to serial chest radiographs (CXRs). We analyzed 391 patients with COVID-19 who underwent serial CXRs during isolation at a residential treatment center (median interval: 3.57 days; range: 1.73–5.56 days). Patients were categorized into two groups: the improved group (n = 309), who completed the standard 7-day quarantine, and the deteriorated group (n = 82), who showed worsening symptoms, vital signs, or CXR findings. Using DLAD’s consolidation probability scores and gradient-weighted class activation mapping (Grad-CAM)-based localization maps, we quantified the consolidation area through heatmap segmentation. The weighted area was calculated as the sum of the consolidation regions’ areas, with each area weighted by its corresponding probability score. Change rates (Δ) were defined as per-day differences between consecutive measurements. Prediction models were developed using Cox proportional hazards regression and evaluated daily from day 1 to day 7 after the subsequent CXR acquisition. Among the imaging factors, baseline probability and ΔProbability, ΔArea, and ΔWeighted area were identified as prognostic indicators. The multivariate Cox model incorporating baseline probability and ΔWeighted area demonstrated optimal performance (C-index: 0.75, 95% Confidence Interval: 0.68–0.81; integrated calibration index: 0.03), with time-dependent AUROC (Area Under Receiver Operating Curve) values ranging from 0.74 to 0.78 across daily predictions. These findings suggest that the Δparameters of DLAD can aid in predicting short-term clinical outcomes in patients with COVID-19. Full article

Intensity–duration–frequency (IDF) curves, which relate rainfall intensity (i), storm duration (d), and return period (T), are cornerstone tools for planning, designing, and operating hydraulic works. Since Sherman’s pioneering formulation in 1931, many modern implementations have systematically omitted the duration-shifting parameter C, causing predicted intensities to diverge to infinity as $d\to 0$. This mathematical paradox becomes especially problematic under extreme hydrological regimes and convective storms exceeding 300 mm/h, where an accurate curve fit is critical. Here, we first review conventional IDF curve fitting techniques and their limitations. We then introduce IDF-GtzLo, a novel, intuitive formulation that reinstates and calibrates C directly from observed storm statistics, ensuring finite intensities for all durations. Applied to 36 automatic weather stations across Mexico, our method reduces the root mean square error by 23 % compared to the classical model. By eliminating the infinite intensity paradox and improving statistical performance, IDF-GtzLo offers a more reliable foundation for hydrological risk assessment and the design of infrastructure resilient to climate-driven extremes. Full article

Dynamic deformation measurement (DDM) is critical across infrastructure and industrial applications. Among various advanced techniques, multi-camera digital image correlation (MC-DIC) stands out due to its ability to achieve wide-range, full-field, and non-contact 3D DDM by pairing camera subsystems. However, existing MC-DIC methods typically rely on inefficient manual pairing or a simplistic strategy that aggregates all visible cameras for measuring specific object regions, leading to camera over-grouping. These limitations often result in cumbersome system setup and ill-measured deformations. To overcome these challenges, we propose a novel MC-DIC method with pointwise-optimized model-based stereo pairing (MPMC-DIC). By automatically evaluating and selecting camera pairs based on five evaluation factors derived from 3D model and calibrated cameras, the proposed method overcomes the over-grouping problem and achieves high-precision DDM of semi-rigid objects. A Ø5 × 5 cm cylinder experiment demonstrated an accuracy of 0.03 mm for both horizontal and depth displacements in the 0.0–5.0 mm range, and validated strong robustness against cluttered backgrounds using a 2 × 4 camera array. Vibration measurement of a 9 × 15 × 16 cm PC speaker operating at 50 Hz, using eight surrounding cameras capturing 1920 × 1080 images at 400 fps, confirmed the proposed method’s capability to perform wide-range dynamic deformation analysis and its robustness against complex object geometries. Full article

High-power-density electric machines play a key role in decarbonising transportation technologies. A critical component of the movement towards high-performance machines is the structure and manufacture of the windings, as this is the dominant source of machine loss. Manufacturing time is important to the effectiveness of the production line, with equivalent importance to the electromagnetic and thermal characteristics. Edgewise windings are increasingly considered to have high potential to be quickly and automatically manufactured. However, they are rarely studied considering all the aspects, these being electromagnetic, thermal, and manufacturing characteristics. This paper will experimentally assess the performance of edgewise machines compared to a stranded winding machine, covering all the aforementioned aspects. Two edgewise winding types are considered, parallel slot and parallel tooth. Firstly, a baseline 11 kW stranded winding machine will be introduced, then two edgewise type machines are proposed to be compared to the baseline machine. These comparisons will initially be made based on simulated torque and thermal performance, then the manufacturing time and quality are assessed for each of the coil structures, showing the achievable time reduction by using edgewise coil structures. Motorettes are used to validate thermal performance of the structures, which are used to calibrate simulation models and evaluate the performance of a full machine equivalent model. Under the thermal limit condition, it is shown that the edgewise parallel tooth windings can achieve a torque increase of 27.8% compared to stranded and 24% compared to edgewise parallel slot. Full article

Miniaturized pressure sensors fabricated via micro-electro-mechanical systems (MEMSs) technology are ubiquitous in modern applications. However, the massively produced MEMS pressure sensors, prior to being practically used, need to be calibrated one by one to eliminate or minimize nonlinearity and zero drift. This paper presents a systematic design for the testing and calibration process of MEMS-based absolute pressure sensors. Firstly, a numerical analysis is carried out using finite element method (FEM) simulation, which verifies the accuracy of the temperature control of the physical calibration system. The simulation results reveal a slight non-uniformity of temperature distribution, which is then taken into consideration in the calibration algorithm. Secondly, deploying a home-made calibration system, the MEMS pressure sensors are tested automatically and rapidly. The experimental results show that each batch, which consists of nine sensors, can be calibrated in 80 min. The linearity and temperature coefficient (TC) of the pressure sensors are reduced from 46.5% full-scale (FS) and −1.35 × 10⁻⁴ V·K⁻¹ to 1.5% FS and −8.8 × 10⁻⁷ V·K⁻¹. Full article

In loop-powered 4–20 mA transmitter systems, sensors like temperature, pressure, flow, and gas sensors are chosen based on specific application requirements. These systems are widely adopted in high-precision measurement scenarios, including industrial automation, process control, and environmental monitoring. The transmitter requires a high-performance analog front end (AFE) for precise amplification and signal conditioning. This paper presents a low-noise instrumentation amplifier (IA) for high-precision transmitter front ends, featuring a Successive Approximation Register (SAR)-assisted offset calibration architecture. The proposed structure integrates a chopper current-feedback instrumentation amplifier (CFIA) with an automatic offset calibration loop (AOCL), significantly suppressing internal offset errors and enabling high-accuracy signal acquisition under stringent power and environmental temperature constraints. The designed amplifier provides four selectable gain settings, covering a range from ×32 to ×256. Fabricated in a 0.18 μm CMOS process, the CFIA operates at a 1.8 V supply voltage, consumes a static current of 182 μA, and achieves an input-referred noise as low as 20.28 nV/√Hz at 1 kHz, with a common-mode rejection ratio (CMRR) up to 122 dB and a power-supply rejection ratio (PSRR) up to 117 dB. Experimental results demonstrate that the proposed amplifier exhibits excellent performance in terms of input-referred noise, offset voltage, PSRR, and CMRR, making it well-suited for front-end detection in field instruments that require direct interfacing with measured media. Full article

Ultrasound probe calibration is crucial for precise spatial mapping in ultrasound-guided surgical navigation and free-hand 3D ultrasound imaging as it establishes the rigid-body transformation between the ultrasound image plane and an external tracking sensor. However, the existing methods often rely on manual feature point selection and exhibit limited robustness to outliers, resulting in reduced accuracy, reproducibility, and efficiency. To address these limitations, we propose a fully automated calibration framework that leverages the geometric priors of an N-wire phantom to achieve reliable recognition. The method incorporates a robust feature point extraction algorithm and integrates a hybrid outlier rejection strategy based on the Random Sample Consensus (RANSAC) algorithm. The experimental evaluations demonstrate sub-millimeter accuracy (<0.6 mm) across varying imaging depths, with the calibration process completed in under two minutes and exhibiting high repeatability. These results suggest that the proposed framework provides a robust, accurate, and time-efficient solution for ultrasound probe calibration, with strong potential for clinical integration. Full article

This article provides a critical review of conceptually different approaches to automatic and transformer-based automatic item generation. Based on a discussion of the current challenges that have arisen due to changes in the use of psychometric tests in recent decades, we outline the requirements that these approaches should ideally fulfill. Subsequently, each approach is examined individually to determine the extent to which it can contribute to meeting the challenges. In doing so, we will focus on the cost savings during the actual item construction phase, the extent to which they may contribute to enhancing test validity, and potential cost savings in the item calibration phase due to either a reduction in the sample size required for item calibration or a reduction in the item loss due to insufficient psychometric characteristics. In addition, the article also aims to outline common recurring themes across these conceptually different approaches and outline areas within each approach that warrant further scientific research. Full article

Educational assessment relies on well-constructed test items to measure student learning accurately, yet traditional item development is time-consuming and demands specialized psychometric expertise. Automatic item generation (AIG) offers template-based scalability, and recent large language model (LLM) advances promise to democratize item creation. However, fully automated approaches risk introducing factual errors, bias, and uneven difficulty. To address these challenges, we propose and evaluate a hybrid human-in-the-loop (HITL) framework for AIG that combines psychometric rigor with the linguistic flexibility of LLMs. In a Spring 2025 case study at Franklin University Switzerland, the instructor collaborated with ChatGPT (o4-mini-high) to generate parallel exam variants for two undergraduate business courses: Quantitative Reasoning and Data Mining. The instructor began by defining “radical” and “incidental” parameters to guide the model. Through iterative cycles of prompt, review, and refinement, the instructor validated content accuracy, calibrated difficulty, and mitigated bias. All interactions (including prompt templates, AI outputs, and human edits) were systematically documented, creating a transparent audit trail. Our findings demonstrate that a HITL approach to AIG can produce diverse, psychometrically equivalent exam forms with reduced development time, while preserving item validity and fairness, and potentially reducing cheating. This offers a replicable pathway for harnessing LLMs in educational measurement without sacrificing quality, equity, or accountability. Full article

During automated packaging of white tea, uneven tea pile thickness leads to reduced weighing accuracy, while traditional experimental methods struggle to reveal the underlying particle flow mechanisms, hindering equipment optimization. Addressing the lack of discrete element method (DEM) parameters for Baihao Yinzhen tea, this study calibrates its DEM parameters based on the DEM approach, providing input for virtual commissioning of packaging machinery. Through physical experiments, the static friction coefficient (0.546), restitution coefficient (0.326), and rolling friction coefficient (0.133) between tea leaves and steel plates were determined. A three-dimensional DEM model of tea leaves was established using slicing techniques and the multi-sphere aggregation method. The steepest-ascent method and Box–Behnken design were employed to optimize the simulation parameters, resulting in the following optimal parameter combination: inter-particle restitution coefficient (0.16), static friction coefficient (0.14), and rolling friction coefficient (0.15). Validation simulations demonstrated that the mean angle of repose of tea leaves under the optimized parameter combination was 22.51°, with a relative error of only 1.29% compared to the actual experimental result of 22.80°. The calibrated parameters can be directly applied to the simulation of the feeding system in white tea automatic packaging machines, enabling optimization of vibration parameters through prediction of pile behavior, thereby reducing weighing errors. Full article

The response time of Artificial Neural Network (ANN) inference is critical in embedded systems processing sensor data close to the source. This is particularly important in applications such as predictive maintenance, which rely on timely state change predictions. This study enables estimation of model response times based on the underlying platform, highlighting the importance of benchmarking generic ANN applications on edge devices. We analyze the impact of network parameters, activation functions, and single- versus multi-threading on response times. Additionally, potential hardware-related influences, such as clock rate variances, are discussed. The results underline the complexity of task partitioning and scheduling strategies, stressing the need for precise parameter coordination to optimise performance across platforms. This study shows that cutting-edge frameworks do not necessarily perform the required operations automatically for all configurations, which may negatively impact performance. This paper further investigates the influence of network structure on model calibration, quantified using the Expected Calibration Error (ECE), and the limits of potential optimisation opportunities. It also examines the effects of model conversion to Tensorflow Lite (TFLite), highlighting the necessity of considering both performance and calibration when deploying models on embedded systems. Full article

A critical gap exists between coastal communities’ need for accessible flood risk assessment tools and the availability of sophisticated modeling, which remains limited by technical barriers and computational demands. This study introduces three key innovations through Coastal Defense Pro: (1) the first operational web-based AI ensemble for coastal flood risk assessment integrating real-time multi-agency data, (2) an automated regional calibration system that corrects systematic model biases through machine learning, and (3) browser-accessible implementation of research-grade modeling previously requiring specialized computational resources. The system combines Bayesian neural networks with optional LSTM and attention-based models, implementing automatic regional calibration and multi-source elevation consensus through a modular Python architecture. Real-time API integration achieves >99% system uptime with sub-3-second response times via intelligent caching. Validation against Hurricane Isabel (2003) demonstrates correction from 197% overprediction (6.92 m predicted vs. 2.33 m observed) to accurate prediction through automated identification of a Chesapeake Bay-specific reduction factor of 0.337. Comprehensive validation against 15 major storms (1992–2024) shows substantial improvement over standard methods (RMSE = 0.436 m vs. 2.267 m; R² = 0.934 vs. −0.786). Economic assessment using NACCS fragility curves demonstrates 12.7-year payback periods for flood protection investments. The open-source Streamlit implementation democratizes access to research-grade risk assessment, transforming months-long specialist analyses into immediate browser-based tools without compromising scientific rigor. Full article

This paper presents a novel design for precise vertical landing of drones based on the detection of three phase shifts in the range of ±180°. The design has three inputs to which the signal transmitted from an oscillator located at the landing point arrives with different delays. The circuit increases the aerial tracking volume relative to that achieved by detectors with theoretical unambiguous detection ranges of ±90°. The phase shift measurement circuit uses an analog phase detector (mixer), detecting a maximum range of ±90°and a double multiplication of the input signals, in phase and phase-shifted, without the need to fulfill the quadrature condition. The calibration procedure, phase detector curve modeling, and calculation of the input signal phase shift are significantly simplified by the use of an automatic gain control on each branch, dwhich keeps input amplitudes to the analog phase detectors constant. A simple program to determine phase shifts and guidance instructions is proposed, which could be integrated into the same flight control platform, thus avoiding the need to add additional processing components. A prototype has been manufactured in C band to explain the details of the procedure design. The circuit uses commercial circuits and microstrip technology, avoiding the crossing of lines by means of switches, which allows the design topology to be extrapolated to much higher frequencies. Calibration and measurements at 5.3 GHz show a dynamic range greater than 50 dB and a non-ambiguous detection range of ±180°. These specifications would allow one to track the drone during the landing maneuver in an inverted cone formed by a surface with an 11 m radius at 10 m high and the landing point, when 4 cm between RF inputs is considered. The errors of the phase shifts used in the landing maneuver are less than ±3°, which translates into 1.7% losses over the detector theoretical range in the worst case. The circuit has a frequency bandwidth of 4.8 GHz to 5.6 GHz, considering a 3 dB variation in the input power when the AGC is limiting the output signal to 0 dBm at the circuit reference point of each branch. In addition, the evolution of phases in the landing maneuver is shown by means of a small simulation program in which the drone trajectory is inside and outside the tracking range of ±180°. Full article

Peak detection is a fundamental task in spectral and time-series data analysis across diverse scientific and engineering disciplines, yet traditional approaches are highly sensitive to the choice of algorithm parameters, complicating reliable and consistent interpretation. Triggered by the requirement for the energy calibration for the 128 detectors of the PI3SO gamma ray scanner, we introduce a versatile methodology inspired by concepts from persistent homology, extending the traditional notion of persistence to a multi-parameter setting. Our approach systematically explores the space defined by multiple detection parameters and quantifies peak robustness through the hyper-volume in the parameter space where each peak is consistently identified. This volumetric multi-parameter persistence (VM-PP) measure enables robust peak ranking and significantly reduces the sensitivity of detection outcomes to individual parameter selection, demonstrating utility across simulated and experimental spectral datasets. Extensive validation reveals that this method reliably differentiates genuine peaks from noise-induced fluctuations under diverse noise conditions, proving effective in practical spectroscopic calibration scenarios. This framework, general by design, can be readily adapted to diverse signal-processing applications, enhancing interpretability and reliability in complex feature-detection tasks. Full article