Automated surface inspection of balsa wood panels is challenging because defects may be small, elongated, weakly contrasted, or visually similar to natural grain patterns. This study proposes YOLOv11–BiFPN–DAAF, an enhanced object-detection architecture that combines bidirectional multi-scale feature fusion with adaptive dual-attention feature refinement. The task was formulated as single-class detection, with all anomalous surface regions labeled as Defect. The dataset comprised 508 manually annotated RGB images, including independent internal and external production test sets. A preliminary screening identified YOLOv11-m512 as the reference configuration, followed by a controlled 2×2 factorial ablation comprising the baseline, BiFPN, DAAF, and their combined integration. Each configuration was trained using five independent random seeds under identical experimental conditions. On the validation set, the combined architecture achieved a precision of 0.893±0.004, recall of 0.848±0.006, mAP@0.5 of 0.897±0.004, and mAP@0.5:0.95 of 0.389±0.004. Relative to the baseline, the largest improvement was obtained for mAP@0.5:0.95, with a relative gain of 9.89%, indicating improved localization under stricter IoU thresholds. The improvement was retained on the independent internal test set, where the proposed architecture reached mAP@0.5 and mAP@0.5:0.95 values of 0.892±0.005 and 0.384±0.006, respectively. On the external production test set, the corresponding values were 0.865±0.007 and 0.358±0.008, representing absolute improvements of 0.034 and 0.042 over the original YOLOv11 baseline. Under the matched experimental protocol, YOLOv11–BiFPN–DAAF also achieved the highest principal detection metrics among the evaluated representative detectors. Although BiFPN and DAAF introduced a moderate computational overhead, the architecture maintained an inference time of 9.6±0.3 ms per image. These findings support the potential of the proposed architecture for automated balsa wood panel inspection, while broader multi-site and hardware-level validation remains necessary before large-scale industrial deployment.
Accurate multi-object tracking and metric localization support traffic monitoring and cooperative intelligent transportation at complex intersections. This study presents a fixed-camera vision-map fusion framework that addresses two practical difficulties: axis-aligned boxes poorly represent turning vehicles, and unconstrained image-plane tracking can produce physically implausible trajectories. A map-aided frontend first generates candidate detections using improved You Only Look Once version 8 nano (YOLOv8n) horizontal bounding box (HBB) branch and an improved YOLOv8 oriented bounding box (OBB) branch. A high-definition (HD) map selector then retains the candidate geometry consistent with the straight-driving or turning region and converts it into a unified detection record. The selected reference point is projected to the ground plane through an offline-estimated homography, whereas the appearance feature bypasses the homography and is passed directly to the association stage. The tracking backend uses a 12-dimensional joint image/metric state, symmetric central-difference evaluations of the process and measurement functions, appearance-motion association, and a feasible-road projection derived from HD-map lane polygons. On the evaluated public sequences, the complete configuration achieved a multiple object tracking accuracy (MOTA) of 74.5%, an identification F1 score (IDF1) of 82.6%, 614 identity switches, and a throughput of 26.8 frames per second (FPS) on an RTX 4090 workstation. In a descriptive Vehicle-in-the-Loop case study involving one instrumented vehicle at one intersection, the overall localization mean absolute error (MAE) was 0.180 m, compared with 0.208 m for the baseline end-to-end configuration. These results indicate the feasibility of combining branch-specific vehicle geometry with map-constrained tracking; controlled same-detector comparisons, repeated multi-vehicle trials, and embedded-device latency and power profiling remain necessary for broader claims.
Type I diabetes mellitus (T1DM) is a chronic metabolic disease resulting from insufficient insulin secretion into the bloodstream‚ causing elevated blood glucose concentrations to dangerous levels․ Automated regulation of blood glucose levels in T1DM can be modeled as a nonlinear‚ uncertain‚ and disturbance-affected closed-loop control process with a time delay‚ time-varying insulin sensitivity, and imperfect glucose measurements. This paper presents the design, multi-objective tuning, and robustness evaluation of a fuzzy logic controller (FLC) for automated insulin-infusion regulation. The proposed FLC uses the glucose tracking error and its time derivative as feedback signals to determine the required insulin control action and maintain glucose within the desired range of (70–160 mg/dL). The controller parameters are optimized using the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to address three competing control objectives: minimizing hypoglycemia risk, minimizing hyperglycemia risk, and reducing total insulin usage. The resulting Pareto-optimal solutions provide a set of trade-off controller designs for decision-makers based on safety, performance, and insulin-efficiency requirements. The robustness of the proposed automated control framework is evaluated under challenging operating conditions, including elevated initial glucose levels, model-parameter uncertainties, external disturbances, variations in insulin sensitivity, distorted glucose measurements, and delayed insulin infusion. A comparative study with a linear quadratic regulator-based controller (LQRC) is conducted as a benchmark. Simulation results demonstrate that the optimized FLC provides superior closed-loop performance and stronger robustness than the LQRC across all tested scenarios. The proposed fuzzy-control framework, therefore, offers a promising automation-based strategy for resilient glucose regulation under uncertainty, measurement imperfections, and actuation delays.
In Industry 4.0 manufacturing, predictive maintenance has emerged as a key focus to ensure the reliability of operations, minimize downtime, and contribute to equipment safety. This study proposes a predictive maintenance framework for additive manufacturing systems (AMSs) using an explainable artificial intelligence (XAI) and machine learning (ML) approach, which is supported by real-time multi-sensor monitoring and is synchronized with a digital twin architecture. A heterogeneous dataset of over 10,000 observations and 13 sensor attributes was gathered from a sensing architecture with ESP32 cameras designed to handle heterogeneous signals from thermal, environmental, mechanical and safety-related sensors. Domain-aware feature engineering was done to gain insights into operation indicators such as temperature instability, vibration degradation, smoke risk, humidity anomalies and aggregated maintenance risk scores. Multiple predictive models, such as the Random Forest model, XGBoost model, Logistic Regression model, K-Nearest Neighbours classifier, Multi-Layer Perceptron model, and ARIMA forecast model, were comparatively assessed under highly imbalanced maintenance conditions. The results showed that ensemble learning methods, especially XGBoost and MLP, had better recall and ROC-AUC for fault detection of maintenance-critical problems. SHAP and LIME analyses then showed that a number of physically meaningful indicators, including thermal instability, vibration anomalies, smoke-related features, and aggregated risk scores, have a significant impact on maintenance predictions and hence provide better operational interpretability and maintenance transparency.
Accurate monitoring of atmospheric ammonia (NH3) is important for air-quality assessment and sustainable agriculture, but available datasets differ in spatial resolution, temporal coverage, units, and physical meaning. This study presents a feature-based machine-learning framework with train–test leakage-controlled preprocessing to evaluate relative NH3 classification consistency across three datasets over China: CAMS GEI, CAMS EAC4, and MEIC. Statistical descriptors were extracted for four spatial–temporal cases: Province–Year, Zone–Year, Province–Season–Year, and Zone–Season–Year. Six classifiers were evaluated using chronological testing and cross-validation. CAMS GEI provided the broadest spatial–temporal coverage, while MEIC showed comparatively stable classifier behavior. The best-performing ensemble models commonly achieved accuracies between 0.97 and 0.99 in the main province-level cases. Mean Absolute Value (MAV) was the leading feature in several province-level analyses, with a maximum reported contribution of 64.26%. These scores describe the separability of threshold-derived reference classes and should not be interpreted as an independent physical prediction of NH3 from external atmospheric drivers.
In aerospace manufacturing, validating heavy-duty automated production tooling and Ground Support Equipment (GSE) traditionally requires costly and time-consuming physical proof load testing. This study proposes a novel Virtual Certification framework that utilizes a high-fidelity Multiphysical Digital Twin driven by a Software-in-the-Loop (SiL) co-simulation architecture (integrating Siemens NX MCD, SIMIT, and TIA Portal) to retroactively diagnose mechanical failures and virtually validate design modifications prior to physical manufacturing. The dual-focus methodology is rigorously applied to a physical case study: an over-constrained (hyperstatic) 4-point aerospace lifting system designed for a 26.48 kN fuselage section that suffered a catastrophic mechanical stall during a 39.24 kN physical proof load verification. While conventional static dimensioning models erroneously predicted a nominal drive torque of only 4.56 Nm, the high-fidelity dynamic twin (incorporating a non-linear exponential Stribeck friction model) calculated the transient mechanical resistance causing the stall, capturing a peak load of 46.2 Nm at the motor shaft. The SiL co-simulation revealed that the rigid positional synchronization logic enforced by the PLC inadvertently amplified localized boundary friction, driving the actuators beyond their rated 6.4 Nm capacity. Based on this forensic diagnosis, a remedial powertrain featuring an 8.0 Nm stepper motor coupled with a 16:1 planetary gearbox was integrated and virtually certified. The framework confirmed that the upgraded architecture successfully attenuated the hyperstatic resistance, reflecting a peak load of only 3.0 Nm at the motor shaft and guaranteeing a stable Safety Factor of 2.66. By bridging the reality gap without iterative physical prototyping, this framework establishes a scalable, “First-Time-Right” validation paradigm for multi-point automated manufacturing mechanisms.
Brinjal (Solanum melongena L.) is an important vegetable crop worldwide, but its cultivation faces challenges from pests, diseases, and variable environmental conditions that negatively affect quality and yield. Accurate fruit detection in natural field conditions is essential for yield estimation and perception module of automated harvesting, but existing deep learning techniques often require substantial computational support, restricting their deployment on edge devices. This study addresses this gap by evaluating the Faster Objects, More Objects (FOMO) model—a lightweight architecture for resource-constrained platforms—for brinjal detection under diverse field conditions. A dataset of 1500 images was captured under varying illumination and growth stages, annotated using a bounding box-based approach, and used to train FOMO models with 25, 50, and 100 epochs via transfer learning. Post-training quantization to INT8 format was applied to assess improvements in computational efficiency. The Float32 model achieved a precision of 0.827, recall of 0.915, and F1-score of 0.869 at 100 epochs. The INT8-quantized model maintained comparable accuracy (precision 0.829, recall 0.908, F1-score 0.866) while reducing model size by 63.33% (from 0.30 MB to 0.11 MB), inference time by 62.02% (from 65.3 ms to 24.8 ms), and RAM usage by 73.02% (from 887.2 KB to 239.4 KB). These results demonstrate that FOMO combined with INT8 quantization provides an efficient, accurate solution for real-time brinjal detection on edge platforms, supporting the advancement of precision agriculture through intelligent crop monitoring and serving as a perception module for future robotic harvesting systems.
Artificial intelligence (AI) and the Internet of Things (IoT) are converging to create densely sensed, connected, and increasingly automated urban environments. However, research on AI–IoT integration remains dominated by optimisation-centric perspectives that under-specify spatial reasoning, physical actuation, and institutional accountability. This perspective addresses that gap by developing closed-loop urban automation as an analytical lens for understanding how AI–IoT systems move from urban sensing to consequential intervention. Drawing on a narrative, theory-driven synthesis of literature on Urban AI, IoT, Physical AI, spatial intelligence, digital twins, robotics, automation, and governance, the paper differentiates the framework from AIoT, cyber-physical systems, embodied AI, and optimisation-centric smart-city models. It identifies the distinctive conditions of urban automation, proposes six examinable propositions, and outlines implications for planning support, intelligent control, human oversight, and democratic legitimacy.
In this paper, the position–orientation tracking of the Stewart platform is investigated. We focus on the target-oriented tracking task, which usually occurs in the application of spotlights and cameras. Specifically, the mobile plate of the Stewart platform is always oriented to an immobile point while tracking a desired path. Based on the velocity-level kinematics and zeroing neural dynamics (ZND), a kinematic tracking model is proposed for the position–orientation tracking task. In addition, a robust ZND (RZND) model is further proposed against two kinds of disturbances. Theoretical analyses are presented to show the convergence properties of the proposed models. Discrete-time algorithms of the models are also developed for the convenient implementation. According to the comparative simulations, both the ZND and RZND algorithms accomplish the position–orientation tracking task without the disturbance, and the disturbance-suppression capability of the RZND algorithm is substantiated.
Occupancy map merging in dynamic environments is challenging because moving objects introduce time-varying disturbances that degrade registration accuracy and structural consistency. This paper proposes a reliability-aware map-merging method that explicitly models temporal validity and probabilistic reliability in occupancy grid maps. The method constructs two complementary maps: a trajectory-guided temporal reliability map that reflects the recency and persistence of cell observations and a probabilistic reliability map that refines detected candidate regions based on temporal persistence and spatial reliability criteria. By suppressing transient clutter before registration, the proposed approach focuses alignment on structurally stable regions and improves merging robustness. Experiments in simulation and real-world indoor environments demonstrate clear improvements over a conventional feature-based baseline, substantially reducing both translation and rotation errors. These results show that incorporating temporal validity and probabilistic reliability can improve occupancy map merging under dynamic conditions.
The present article discusses the application of genetic algorithms (GA) for solving multi-criteria optimization (MCO) problems in underground mining. It has been demonstrated that GAs are highly effective in identifying Pareto-optimal solutions in scenarios involving multiple conflicting criteria, specifically the simultaneous minimization of equipment failure rate, energy consumption, and repair costs. The article presents the main approaches to solving MCO problems, a brief overview of the most popular algorithms, such as NSGA-II and SPEA2, and their improved versions. The proposed algorithm, implemented in Python 3.11 using the DEAP library, incorporates adaptive crossover, enhanced diversity preservation, and problem-specific initialization. Quantitative analysis shows that the proposed algorithm achieves a Hypervolume Indicator of 0.796, representing a 7.2% improvement over standard SPEA2, with an 18.3% reduction in Inverted Generational Distance (IGD), indicating superior convergence to the true Pareto front. The algorithm identifies optimal trade-offs between conflicting objectives—for example, a 15% reduction in energy consumption correlates with a 10% increase in failure rate—providing decision-makers with quantified insights for operational planning. The novel idea is the use of an adaptive crossover strategy, a composite diversity maintenance technique, and application-specific initialization—all of which have not been used before for optimizing underground mining machinery. A visual analysis of the results, employing a graphical representation of the Pareto front, confirmed that the proposed approach enables experts to make informed decisions based on production priorities.
This research presents an integrated framework for operational planning of low-power robotic agricultural systems, which combines digital twins, uncertainty modeling with triangular fuzzy numbers, and multi-objective optimization in a coherent structure. The goal is to balance energy consumption, carbon emissions, operational delay, and crop yield under variable and uncertain field conditions. The proposed framework was evaluated using real and simulated data, various operational scenarios, and comparative analyses. The results showed that this approach reduced energy consumption from 248.6 to 191.5 kWh and carbon emissions from 132.4 kg CO2 to 96.8 kg CO2, while increasing crop yield from 148.7 to 178.4 kg/day, compared to the deterministic baseline model. Also, the use of digital twins improved the quality of decision-making in different scenarios by about 6 to 7 percent, and fuzzy modeling significantly increased the stability of results at higher levels of uncertainty. The findings show that the proposed framework can be an effective tool for sustainable, smart, and energy-efficient agriculture.
Computer numerical control (CNC) machining process simulation is increasingly central to intelligent manufacturing, yet its deployment in brownfield environments remains constrained by legacy controllers, heterogeneous data semantics, limited computational resources, and rising cybersecurity requirements. While digital twins (DTs), artificial intelligence (AI), and multi-physics simulation have matured conceptually, practical adoption, particularly among small and medium-sized enterprises (SMEs), continues to lag behind theoretical capability. This paper presents a PRISMA-guided systematic review of peer-reviewed literature, standards, and industrial reports published between 2019 and 2025, focusing on CNC machining simulation, digital twin architectures, interoperability standards, and intelligent optimisation under brownfield constraints. Rather than proposing new simulation algorithms, the review synthesises fragmented evidence into a deployable, standards-aligned integration perspective. The review consolidates prior work into a seven-layer architecture grounded in ISO 23247, explicitly separating sensing, communication, digital twin entities, analytics, and human–machine interaction. It derives practical decision rules for middleware selection, edge-cloud compute placement under latency constraints, and modelling strategy selection, ranging from mechanistic and finite-element methods to hybrid reduced-order and machine-learning surrogates. An SME-oriented implementation and validation roadmap links staged retrofitting to measurable operational indicators, including overall equipment effectiveness, first-pass yield, downtime, cycle time, and energy intensity.
AI-enabled cyber-physical systems (CPSs) are increasingly deployed in public governance contexts where they sense human populations, infer classifications or risks, and trigger interventions that can shape liberty, equality, and access to essential services. In these deployments, governance failures often arise not only from model error but from systems-level interactions across data generation, model updates, organizational practices, and downstream actuation. This paper introduces a Risk-Rights-Rules (3R) architecture that treats fundamental rights and legal rules as enforceable constraints on the sensing-inference-actuation loop, rather than as external ethical aspirations. Building on established risk-management baselines and safety engineering practice, we specify a testable assurance object, a structured 3R assurance case, that links rights claims to explicit assumptions, measurable evidence, and accountable control points across the lifecycle. The approach is designed to reduce "legitimacy drift" in stochastic decision pipelines by making uncertainty, demographic error, contestability, and procurement leverage auditable at the system level. The result is a governance blueprint for high-consequence public-sector AI deployments for governance failures, which is both technically robust and institutionally defensible.
The demands of competitiveness in global markets require the integration of Industry 4.0 (I4.0) digital technologies for any manufacturing company, regardless of size. Industrial operations require complete supply chain visibility to ensure data protection and authenticity throughout the process. This document presents a distributed architecture based on RAMI 4.0, designed for product traceability in industrial environments. It integrates automation tools, IIoT communication, cloud storage, artificial intelligence, and secure data transmission using encrypted communication protocols. The system consists of a hybrid architecture; only the first, lower-level layer corresponds to a simulated manufacturing plant with deterministic and stochastic dynamics within the production line. In the second part, the middle and upper layers are implemented, where plant data is transmitted to a cloud instance, stored in a PostgreSQL database, and subsequently analyzed using automated scripts. Reporting capabilities are incorporated with ChatGPT-3.5 Turbo, and visualization is provided through Odoo. Experimental tests demonstrated an average end-to-end communication latency of less than 200 ms, a packet loss rate of 2.67%, and 100% reliability in verifying requested reports when using the cognitive computing service. Furthermore, the results of the systematic vulnerability identification process for the architecture show a significant reduction in overall risk for most assets, with a predominant shift from high or moderate to low or moderate. The proposed architecture is validated in a simulated industrial environment under controlled conditions, demonstrating its viability as a prototype rather than as a fully implemented industrial solution.
The rapid integration of electric vehicles (EVs) into transportation systems and power grids has significantly increased the complexity of optimization challenges related to routing, charging coordination, scheduling, and energy management. Despite significant research growth, the field remains conceptually fragmented, lacking a unified framework to systematically organize Electric Vehicle Optimization Problems (EVOPs). To address this gap, this study presents a systematic review of 144 peer-reviewed articles published between 2011 and January 2025 and proposes a structured EVOP taxonomy based on problem characteristics and dominant decision variables. The analysis examines mathematical formulations, solution methodologies, and emerging research trends. The results indicate the predominance of metaheuristic methods, while exact techniques are mainly limited to small-scale problems. Additionally, there is a growing trend toward multi-objective and stochastic models that incorporate uncertainty and dynamic decision-making environments. However, challenges remain regarding large-scale validation, standardized benchmarking, and integrated multi-domain modeling. The proposed taxonomy provides a coherent framework that facilitates comparison across optimization domains and supports the development of scalable and intelligent EV management systems.
To develop secure, fast, and interoperable smart substations, it is vital to understand the current situation and potential future directions of the technologies involved. This study presents the evolution and state of the art of the Generic Object Oriented Substation Event (GOOSE) communication protocol, defined by the International Electrotechnical Commission (IEC) 61850 standard. A Systematic Literature Review (SLR) was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol. This included journal articles published from 2004 to 2025 and conference papers from 2020 to 2025, written in English within Engineering. Only studies primarily focusing on GOOSE, citing it at least ten times, and indexed in the Scopus, IEEE Xplore, and Web of Science databases were included. The quantitative analysis used SciMAT software, complemented by a qualitative analysis. Due to the bibliometric and thematic nature of this review, potential biases were considered at the review level rather than by applying a formal study-level risk-of-bias tool. The final analysis comprised 82 journal articles and 84 conference papers. The results offer a comprehensive mapping of GOOSE research evolution, identify nine main challenges and limitations from the last 22 years, and highlight current research directions. The literature reveals methodological heterogeneity, a predominance of simulation-based approaches, and limited large-scale empirical validation.
The object of the research is the information interaction processes between the components of a simulation model of a PID controller based on Digital Twin technology. The problem addressed lies in the need to extend the functionality of various models when they are integrated into real control systems. The aim of the study is to develop a simulation model of a PID controller for electric-drive frequency-based control systems using Digital Twin technology. A concept for constructing a simulation model using unified hardware-software tools from Simatic S7 and Digital Twin technology is proposed. In this approach, virtual components of the simulation model are configured, parameterized, and programmed within the same engineering environment as the real ones. Projects developed based on simulation results of PID controllers provide the foundation for their implementation on real Simatic S7 hardware. The simulation model provides for integration and interaction of fully virtual components, including PLC, frequency converter, electric drive, SCADA, and communication environment. Procedures for parameterizing monitoring tools and for the automatic tuning of PID controller parameters according to the chosen strategy were implemented, which enabled a clear graphical evaluation of transient processes under different operating modes of the simulation model. The response of the PID controller to periodic and random disturbance signals within up to100% of the control range was tested.
The article addresses the challenges of modernizing Kazakhstan’s railway infrastructure under conditions of technological dependence on foreign automation systems and obsolete relay-based equipment. These factors pose significant risks to economic and information security and limit the throughput capacity of level crossings. A digital system, KZ-DALCS-AI, is proposed, based on a multi-level safety architecture and the integration of artificial intelligence into monitoring and control processes. A key component is an obstacle detection and classification algorithm that considers object types (vehicles, humans and animals, foreign objects, and environmental factors) and enables intelligent real-time decision making using the KZ-ODC-AI controller with data from video surveillance, microwave sensors, and inductive loops. The system architecture, operational logic, and level crossing control algorithm are developed, including optimization of closing time by minimizing the deviation between calculated and actual values. The results of the performed calculations confirm the effectiveness of the proposed notification algorithm, ensuring the required level of safety while reducing unnecessary delays for road traffic. The implementation of the system improves throughput, reduces operational costs, enhances reliability, and minimizes the impact of the human factor.
Ensuring safety in autonomous vehicles (AVs) requires predictive control methods that can handle dynamic constraints, uncertain interactions, and real-time decision making. This review examines safety-oriented model predictive control (MPC) for AVs using a PRISMA-guided screening process. From 363 records published between January 2015 and March 2026, 101 peer-reviewed studies were selected for qualitative synthesis. The literature is organized into three domains: collision avoidance and risk mitigation, trajectory tracking and path following, and intersection and coordination tasks. Across these domains, MPC has evolved from nominal tracking and geometric avoidance toward risk-aware, robust, hierarchical, and learning-enhanced formulations. Unlike broader reviews on autonomous driving control, this review focuses specifically on safety-oriented MPC and compares the reviewed literature in terms of safety mechanisms, uncertainty treatment, validation practice, computational feasibility, and deployment limitations. The review shows that MPC remains one of the most versatile frameworks for AV safety, but the evidence base is weakened by heavy reliance on simulation, inconsistent safety metrics, limited validation under uncertainty, and uneven treatment of computational feasibility. The most promising directions are hybrid architectures that combine model-based safety guarantees with uncertainty-aware prediction, learning-assisted adaptation, and scalable coordination mechanisms.
