Keywords: Adaptive control, System identification, Marine control
Abstract: We consider a dynamic positioning model of a surface vehicle, which is affected by bounded non-vanishing perturbation with non-zero mean. The main goal of this study is to estimate such model parameters with zero parametric error under the condition that the model is used in the closed loop and, therefore, the disturbance and regressor of the system are not independent. In order to achieve it, (i) the vessel model is parametrized into the perturbed linear regression form with both measurable regressor and regressand, (ii) a procedure of instrumental variables based dynamic regressor extension and mixing that has been recently proposed by the authors is applied. As a result, a set of scalar regression equations with respect to the vessel parameters with asymptotically vanishing perturbation is obtained, and consequently, such parameters are exactly identified in the asymptotic sense. Both simulation and experiment with a real surface vessel illustrate the presented theoretical result.

Keywords: Automotive control, Fuzzy logic and fuzzy control, Networked systems
Abstract: This work concerns the event-triggered direct yaw control for vehicle stability and handling improvement during critical maneuver. To improve the robustness of the controller two aspects have been considered: nonlinearities of cornering forces and road friction variation. A sampled event-triggered scheme is proposed in order to save communication and energy resources. LMI-based sufficient conditions are designed for reference model tracking fuzzy control with asynchronous premise variables. The efficiency of the proposed method is illustrated on Carsim software.

Keywords: Automotive control, Modelling and simulation, Adaptive control
Abstract: Vehicle dynamic control plays a crucial role in preventing accidents by reducing the difference between the expected and measured vehicle response. Torque vectoring control is a contemporary control method used in vehicle dynamic control to enhance the traction performance of vehicles. In this study, a gain-scheduled PID-based torque vectoring controller is proposed to control a nonlinear, multi-degree-of-freedom vehicle model with three electric motors during turning maneuvers. To this end, twelve different maneuvers that the vehicle may encounter during driving are determined, and optimal coefficients that minimize the total squared error performance criterion are found for each maneuver. The coefficient surfaces are obtained for each controller coefficient using the cubic spline interpolation method. In simulation studies, the proposed gain-scheduled PID controller is tested under various maneuver changes, and it is observed that the performance of the control system significantly improves compared to the conventional PID controller.

Keywords: Intelligent control systems, Adaptive control, Automotive control
Abstract: This paper presents a novel model-free static output feedback (SOF) control strategy for continuous-time systems, leveraging integral reinforcement learning (IRL) within an off-policy framework. In many practical scenarios, full-state measurements are unavailable, requiring control based solely on sensor outputs. The proposed method addresses this challenge by utilizing only input-output data, avoiding the need for state estimators or observers. The approach offers a practical balance of simplicity, stability, and performance, making it a more accessible alternative than dynamic output feedback methods. The proposed algorithm is adaptive and suits real-world applications involving uncertain or unknown system dynamics. Simulation results demonstrate the method's effectiveness, with comparative analysis showing its better performance over existing model-free approaches.

Keywords: Intelligent transportation systems, System identification, Computational methods
Abstract: Estimating Origin-Destination (OD) matrices is a fundamental problem in transportation planning, as they provide critical insights into travel demand and traffic flow distribution. Traditional methods rely on traffic surveys, vehicle tracking, and network tomography techniques, but these approaches often suffer from high costs, limited data availability, and significant estimation uncertainties. In this paper, we present a novel approach to OD estimation that leverages joint cumulants and bootstrapping techniques to improve the robustness of OD demand predictions. Unlike previous methodologies that rely on extensive prior information or require full statistical knowledge of network flows, our method operates under realistic constraints where only a subset of flow measurements is available. By estimating the covariance matrix of joint cumulants and applying a generalized least squares (GLS) approach, we systematically reduce estimation errors while ensuring computational efficiency. Simulation results on both synthetic and real-world datasets indicate that our method performs well in terms of accuracy, suggesting its potential usefulness for traffic management applications.

Keywords: Aerospace control, Predictive control, Neural networks
Abstract: Model Predictive Control (MPC) is a promising method for flight control, offering precise stabilization and maneuvering by predicting system behavior using a model of the aircraft dynamics. Essential for these dynamics are the aerodynamic coefficients. While conventional aerodynamic models often do not meet the real-time requirements of flight control applications, neural networks (NN) promise to accurately capture aerodynamic behavior. However, their computational feasibility in real-time MPC remains an active research area. This paper presents a nonlinear Model Predictive Flight Control strategy for a fighter aircraft, where the numerical solution of the MPC problem requires the gradients of the aerodynamic tables. Instead of modeling the aerodynamic coefficients directly with NNs, we propose to use the original look-up tables and only model their derivatives with low-dimensional feedforward NNs. Simulation results of an MPC demonstrate enhanced computational efficiency without sacrificing accuracy, where the NN modeling makes the gradient computation more than three times faster than conventional difference quotient calculations.

Keywords: Autonomous systems, Disturbance rejection, Intelligent transportation systems
Abstract: The estimation and detection of cyber-attacks in Connected and Autonomous Vehicles (CAV) are paramount for enhancing security and advancing vehicle automation. This paper presents a solution for estimating cyber-attacks in Cooperative Adaptive Cruise Controllers (CACC) while following a preceding vehicle. Our proposed method utilizes the derivative of the output measurements to derive a new system that satisfies the observer matching condition (OMC). Then, a High Order Sliding Mode Observer (HOSMO) is used to estimate these derivatives and reconstruct the cyberattack signal. The stability analysis is performed by using the Lyapunov stability theory. Several test scenarios are investigated to demonstrate the effectiveness and validity of the proposed algorithm.

Keywords: Industrial automation, manufacturing, Robotics, Real-time control
Abstract: Free-swinging hangers on overhead conveyors are widely used in the manufacturing industry, such as in assembly lines and paint finishing lines. Manual loading and unloading of hangers can pose health and safety risks to workers and constitute a substantial part of production costs. In this work, we explore automation of the loading and unloading process using a robot manipulator arm and model-based object tracking. Dynamic robot motions are planned in real time to safely approach the swinging hanger and hook and unhook objects, guided by pose estimates from a CAD-based tracker. The proposed solution demonstrates a flexible approach to automating overhead hanger handling, particularly for retrofitting existing conveyor systems.

Keywords: Process control
Abstract: Particulate processes are widely used in different industries. Due to their complexity and intransparency they are often controlled manually and open-loop in practice, with classic automatic control approaches being hard to apply. To close the control loop of a liquid-liquid mixer, we consider Model Predictive Control. The system is described by a Gaussian Process Regression-Model based on experimental data, since the well explored population balance models (PBM) are difficult to identify for a real system and require significant computational effort, making them hard to use in an automatic control scheme. The relevant cost function and weights take into account not only the mixing process, but also the favorable conditions for the subsequent settler.

Keywords: Real-time control, Wireless sensor networks, Optimisation
Abstract: The Internet of Things (IoT) is revolutionizing modern life by connecting everyday objects to the Internet, enabling real-time data collection and improved decision-making. This work proposes a telemonitoring-based system that explores the optimization of energy efficiency. Our focus is on key parameters such as external temperature and domestic activity, with an emphasis on real-time energy consumption management. By leveraging IoT technologies, this study aims to achieve sustainable and efficient resource management while ensuring system reliability.

Keywords: Unmanned systems, Autonomous systems, Robotics
Abstract: This article presents a coverage path planning pipeline for inspection of complex structures in 3D using unmanned aerial vehicles (UAVs). The approach generates candidate view points through random sampling of an input model and computes an optimal inspection path through a subset of the views using an RRT*-based planner. The planner optimizes coverage and flight time while penalizing hard turns, and ensures a collision-free and safe path by dilation of the mesh model. Software-in-the-loop (SITL) simulation results demonstrate the effectiveness of the inspection path planning pipeline in producing collision-free, high-coverage inspection paths for complex models that take the camera and gimbal parameters into account.

Keywords: Fault diagnosis, Complex systems, Marine control
Abstract: Fault diagnosis in marine dual fuel systems is essential due to the complex operating conditions and the potential for significant safety and environmental impacts. This paper uses Long Short-Term Memory Autoencoder (LSTM-AE) for fault detection in the exhaust systems of marine dual fuel engines. The LSTM-AE model is designed to detect anomalies in exhaust valve closing dead time (ECDT) using real historical data. The methodology involves creating input sequences, encoding and decoding them using LSTM layers, and detecting faults based on reconstruction errors. The model is trained on healthy data from Cylinder 3 and tested on other cylinders, achieving a precision of 1.00, recall of 0.82, and F1-score of 0.90. Results showed that the model successfully detects both known and unreported faults, demonstrating its potential for real-time fault monitoring in marine propulsion systems.

Keywords: Fault diagnosis, Industrial automation, manufacturing, Real-time control
Abstract: This paper develops a real-time data-driven approach to fault diagnosis on three three-phase induction motors, offering unmatched scalability, accuracy, and flexibility. To improve feature representation, extensive datasets were gathered under various load circumstances using vibration, temperature, and current sensors. These datasets were then carefully preprocessed and examined. With a real-time classification accuracy of 99.3%, the XGBoost algorithm; which was chosen for its interpretability and robustness, performed better across unbalanced datasets. By fusing advanced machine learning with hardware that has limited resources, this study pushes the limits of predictive maintenance. It creates a scalable platform for implementing intelligent diagnostic solutions across a range of industrial applications in addition to reducing downtime and operational disturbances. The results have the potential to revolutionize companies striving for safer, smarter, and more sustainable operations in the context of Industry 4.0.

Keywords: Fault diagnosis, Networked systems, Signal processing
Abstract: This paper aims to detect and locate a soft fault, i.e. a physical degradation, in unshielded twisted pair cables used for data transmission in a networked control system (NCS). The proposed approach is based on the analysis of the transmission coefficient (TC) in the time domain for both no-fault and faulty situations. A health indicator ratio is used to estimate the severity of the fault, while a residual is defined to localize the position of the fault. Simulation results are presented to illustrate the method.

Keywords: Fault diagnosis, Networked systems
Abstract: In a drinking water network, a telecontrol system must periodically acquire, store and validate sensor data to achieve accurate monitoring in real-time. Sensor data need to be validated before further use to assure the reliability of the results obtained when using them. In real operation, problems affecting the communication system, lack of reliability of sensors, or other inherent errors often arise, generating missing or false data during certain periods of time. These data must be detected and replaced by estimated data. In this paper, a methodology for data validation and reconstruction of sensor data collected from a drinking water network is developed, considering not only spatial models, but also temporal models (time-series of each sensor) and internal models of the several components in the local units (e.g., pumps, valves, flows, levels). The methodology is illustrated by means of its application to flow and level meters of the main Catalonia Regional Water Transport Networks.

Keywords: Fault diagnosis, Neural networks, Nonlinear systems
Abstract: Accurate anomaly detection in wastewater treatment plants (WWTPs) is essential to secure operational efficiency, environmental compliance and public health safety. This paper investigates two Deep Learning approaches employed for detecting mechanical faults injected in the Dissolved Oxygen (DO) sensor of a WWTP, namely Convolutional Neural Network (C-NN) and Long Short-Term Memory Neural Network (LSTM-NN). Experimental results demonstrate the strengths and limitations of each approach, with both achieving good levels of accuracy, precision, recall and F1-score under varying operational conditions. However, when evaluating the values of the performance metrics obtained for the two Deep Learning approaches, it is proved that the LSTM-NN is more suitable for the task at hand. This study provides insights into the application of Neural Networks for improving fault detection accuracy and enhancing the reliability of wastewater treatment plants.

Keywords: Biologically inspired systems, Modelling and simulation, Computational methods
Abstract: A fundamental property of all living cells is their ability to regulate their size while proliferating. Cell proliferation involves a mother cell, growing over time and dividing into daughter cells at an appropriate moment. Cell size, quantified using metrics such as length in rod-shaped bacteria or mass in human cells, remains under robust control across diverse cell types. Fluctuations in size arising from errors in growth and division cycles can be used to characterize the underlying mechanisms of size control and proliferation. In this work, we develop a stochastic dynamical model to capture cell-to-cell size fluctuations within a homogeneous population of proliferating cells. Size control is implemented by modeling division as a stochastic process with a continuous rate that depends nonlinearly on cell size. This framework leads to a system of partial differential equations (PDEs) that describe the time evolution of the joint distribution of key variables, including cell size, cell cycle timer, and added size since birth. By numerically solving these PDEs, we provide insights into the statistical moments of cell size (mean, coefficient of variation, and skewness), which display oscillatory dynamics over time. With access to quantitative single-cell size data, the computational framework developed here offers a powerful tool to infer and elucidate size control mechanisms across diverse cell types.

Keywords: Biologically inspired systems, Neural networks, Marine control
Abstract: This paper presents a bioinspired robotic fish adaptable to diverse applications, integrating artificial intelligence for real-time recognition of marine objects and a biomimetic locomotion system based on a 2 Degrees of Freedom (DoFs) tail. The onboard neural network processes visual data to identify specific underwater objects, such as mussels and clams. At the same time, a real-time communication system transmits findings remotely via a Telegram-based API. Performance evaluation, including accuracy, loss, confusion matrix, and Receiver Operating Characteristic-Area Under the Curve (ROC-AUC) analysis, confirms the model's reliability in classification tasks. Additionally, key operational metrics such as waterproof integrity, response time, manoeuvrability, and regime speed were measured, demonstrating its practicability. The proposed system offers a versatile and cost-effective solution suitable for applications in marine biology, underwater monitoring, and autonomous exploration.

Keywords: System identification, Genetic and evolutionary computation, Marine control
Abstract: Bearings-only Target Motion Analysis (BOTMA) is a challenging problem in signal processing due to its non-linear nature and reliance on noisy angular measurements. Traditional iterative methods, such as Maximum Likelihood Estimation (MLE), are computationally demanding and sensitive to initialization. This paper presents a novel approach integrating Ant Colony optimisation (ACO) with a System-on-Chip Field Programmable Gate Array (SoC FPGA) to efficiently solve the MLE problem in BOTMA. The implementation is deployed on the PYNQ-Z1 platform, featuring an AMD Zynq-7000 SoC FPGA, which combines an ARM processing system with programmable logic. The FPGA design uses the hardware's concurrent processing capabilities to accelerate the ACO algorithm, achieving low-latency and high-throughput performance. The proposed solution is validated through simulations, demonstrating improved execution performance and low power consumption compared to the software model executed on a general-purpose processor. This work highlights the effectiveness of integrating advanced optimisation techniques with an FPGA-based SoC platform, enabling real-time BOTMA applications in resource-constrained environments.

Keywords: Navigation, Industrial automation, manufacturing, Cyber-physical systems
Abstract: Modern logistics operations in warehouses involve humans working in indoor spaces, often filled with hazardous situations. To establish safe and efficient environments and mitigate the risk of collisions or accidents, it is crucial to localize and track human movements. In the preceding decade, advancements in location-based services, including human activity monitoring, have catalyzed an escalating level of interest in the domain of indoor and outdoor positioning and localization technologies. This study presents a tracking system that combines inertial measurement data with an ultra-wideband infrastructure to accurately track and localize individuals within a warehouse environment, delivering real-time alerts to enhance safety. The system has been designed to be self-configurable and is capable of learning most of the necessary parameters online. Its low computational requirements make it suitable for implementation on wearable devices. The effectiveness of the adaptive online algorithms was demonstrated in a real indoor scenario when the tracking system was tested. A system of alerts has been implemented to ensure that persons remain within the designated safe zone.

Keywords: Navigation, Unmanned systems, Aerospace control
Abstract: This paper proposes a novel nonlinear observer to estimate the position of a flying vehicle in waypoint navigation using quadratic measurements. The quadratic measurements are the squared distance to a point source located at the origin. It is shown that the dynamics of the estimation error are described by a nonlinear Verhulst logistc equation. This observation unveils a link between navigation of a vehicle using squared range measurements and population dynamics, which to the best of the authors' knowledge is new. The convergence of the state estimate is studied for 3D motion. Straight and helical paths are studied and conditions for stability of the error dynamics are formulated. Simulations show the peformance of the nonlinear observer.

Keywords: Navigation
Abstract: In this work, a novel reactive, frontier-based exploration scheme is presented. The approach is based on harmonic transformations, which provide a homeomorphism of arbitrary workspaces onto a punctured disk, thus enabling provably safe navigation. Additionally, in contrast to previous reactive methods, the proposed scheme enables choosing any unexplored frontier for further exploration, while preserving full exploration guarantees in finite time. The relevant claims are proven mathematically, while the scheme is validated via synthetic simulation environments, as well as via a real-world experiment with a mobile robot, demonstrating the capabilities and applicability of the method. The method is successful in all tested cases, while also outperforming relevant schemes in computational time.

Keywords: Unmanned systems, Navigation, Optimisation
Abstract: This paper considers a problem of privacy-aware trajectory planning for an Unmanned Aerial Vehicle (UAV). The objective is to navigate the UAV from an initial position to a final position within a given time limit so that the UAV avoids no-fly zones and areas with high privacy violation risk while minimizing a cost function describing the total privacy violation risk. An effective algorithm for UAV trajectory planning is developed, and its global optimality is proved.

Keywords: Neural networks, Image processing, Biomedical engineering
Abstract: Kidney stones, or renal calculi, affect approximately 10% of the global population and can lead to severe complications such as urinary obstruction and kidney failure. We propose a two-stage pipeline for automated detection and segmentation of kidney stones in CT images. The first stage uses a YOLOv11 network trained on a public dataset to identify regions of interest (ROIs). High-quality segmentation masks, generated using the Segment Anything Model (SAM), were added to the same dataset and used to train a U-Net for detailed segmentation. The final pipeline combines YOLO for ROI detection and U-Net for segmentation, achieving strong performance with metrics such as a Dice coefficient of 0.93 and IoU of 0.88. This approach contributes to kidney stone research and provides a scalable framework for nephrolithiasis assessment in clinical applications.

Keywords: Image processing, Industrial automation, manufacturing, Computational intelligence
Abstract: The improvement of flexible manufacturing systems towards sustainable use of raw materials and increased resource efficiency represents a core tenant of Industry 5.0 competitiveness. This can be currently achieved through the adoption and accelerated implementation of state-of-the-art artificial intelligence models in forecasting, anomaly detection and classification applications. Human-centric approaches balance the deployment and implementation models for control and cognition with socially relevant goals for increased resilience. The paper presents and embedded learning application for vision-based defect detection on a five-station connected laboratory flexible manufacturing line. Quantitative results are illustrated and discussed that comparatively benchmark multiple generations of the YOLO real-time object detection model family along with implementation considerations and integration aspects with industrial automation technology.

Keywords: Image processing, Neural networks, Biomedical engineering
Abstract: Accurate classification of foveal avascular zone (FAZ) from optical coherence tomography angiography (OCTA) images is critical for the early detection and management of retinal pathologies such as diabetic retinopathy and myopia. In this study, we propose a Kernel Principal Component Analysis (KPCA) deep ensemble approach for classifying FAZ in OCTA images. Our framework first extracts the regions of interests (ROIs) and then generates deep features from multiple pre-trained models (ResNet50, VGG16, EfficientNetB0 and DenseNet201) to capture diverse, high-level representations of FAZ regions. Then, KPCA is used to fuse these features into a compact, non-linear representation. On top of this fused feature set, is build a deep classifier, a fully connected feedforward neural network (FFN), to differentiate among normal, diabetic and myopic conditions. Experimental results on the FAZID dataset demonstrate outstanding performance, with a training accuracy of 98.58% and a test accuracy of 98.36%. These results highlight the effectiveness of combining Kernel PCA with ensemble learning in capturing subtle yet clinically significant variations within the FAZ region. The proposed approach not only enhances classification accuracy and generalization but also supports more reliable and automated clinical decision support in retinal diagnostics.

Keywords: Image processing, Neural networks, Computational intelligence
Abstract: Real time localisation and classification of Unexploded Ordnance (UXO) can significantly benefit from advanced new model compression and quantization techniques towards embedded deployment on resource constrained fixed or mobile hardware platforms. This can extend the applicability, usefulness and adoption by first responders of such methods in real-world scenarios with significant social and environmental benefits. The proposed methodology considers the emergence of multiple frameworks and tools that have now become available to automate the comparative assessment of state-of-the-art image classification edge AI model. As main results, we present a quantitative evaluation of the robustness-runtime efficiency trade-off for representative CNN-based vision model and a parametrization discussion on a reference public UXO dataset. The approach is validated through deployment and experiments using a reference embedded GPU development board i.e. the Nvidia Xavier NX.

Keywords: Computational intelligence, Computational methods, Neural networks
Abstract: The classification of histopathological images that contain repetitive patterns is a challenge for AI (artificial intelligence) algorithms developed in the last years. Histopathologic diagnosis continues to be the gold standard for cancer diagnosis despite the quick advances in medical research. Therefore, this study proposes to analyse the factors that influence the classification of benign and adenocarcinomas in colon cancer histopathological images, using four different CAE-CNNs (AutoEncoder Convolutional Neural Networks). This paper describes the geometry and tests for each of the following CAE-CNN: (i) symmetric encoder-decoder with bottleneck layers; (ii) asymmetric encoder-decoder with bottleneck layers; (iii) symmetric encoder-decoder without bottleneck layers; (iv) asymmetric encoder-decoder without bottleneck layers. , where only colon histopathological images were classified. The obtained results are very remarkable: an accurate symmetric encoder-decoder with bottleneck layers that achieves 99.2% accuracy on test images.

Keywords: Modelling and simulation, Networked systems, Computing and communications
Abstract: This paper aims to analyses CDMA-GSM systems using Tetcos' NetSim software by modifying various parameters of the network equipment as well as the transmission application. These parameters include buffer size, transmission power of base stations, mobility, speed, and transmission power of mobile stations. In this application, parameters such as codec, encryption, quality of service, and type of service are modified.

Keywords: Fault diagnosis, Nonlinear systems
Abstract: The paper addresses the design of fault detection and isolation filters using homogeneous finite-time state observers. The proposed filter guarantees finite-time convergence of the observation error in the absence of faults, while in the presence of a fault, the residual aligns with a fixed direction associated with the specified fault. The effectiveness of the approach is demonstrated through a simulation example.

Keywords: Fault diagnosis, Power systems and smart grid, Robotics
Abstract: Visibility maps are crucial for autonomous robotic applications such as exploration, path planning, obstacle avoidance, and multi-robot coordination. In the context of electrical transmission infrastructure, automated robotic inspection enhances proactive maintenance, enabling early detection of wear, damage, or faults, thereby improving safety, extending component lifespan, and optimizing maintenance schedules.

In this work, we propose an algorithm to compute optimal visibility locations, enabling a mobile robot to acquire RGB and thermal images for fault detection. Using LiDAR scans, we construct a global 3D map composed of ground structures (represented as a Grid Map) and overground structures (modeled with an Octomap for efficient ray-casting). We apply clustering techniques to identify 3D bounding boxes for electrical components and define suitable source and target points for visibility assessment. By employing a weighted visibility scoring approach, we determine the ground positions that offer the best visibility of each component while ensuring minimal occlusions and adherence to viewing constraints.

The proposed method enables a robot to autonomously navigate to these optimal viewpoints, improving inspection efficiency. By integrating visibility regions across multiple components, the inspection process is further optimized, reducing overall inspection time. Our algorithm has been successfully deployed and tested at an Ultra-High Voltage Center (UHVC) in Greece, demonstrating its effectiveness in real-world conditions.

Keywords: Automotive control, Optimisation, Intelligent transportation systems
Abstract: Eco-driving has emerged as a promising approach to reducing fuel consumption in road vehicles by optimizing driving behavior for enhanced system efficiency. This paper formulates the eco-driving problem within an optimal control framework. Due to the nonlinear dynamics and complex operational constraints, dynamic programming (DP) is employed to solve the optimization problem. To further improve computational efficiency and ensure constraint compliance, we propose a hybrid method that integrates DP with a regression-based algorithm. The proposed approach is validated through co-simulation using Matlab/Simulink and CarSim, demonstrating its effectiveness in achieving fuel-efficient vehicle operation.

Keywords: Optimisation, Predictive control, Prognostics and diagnostics
Abstract: In an industrial context, monitoring machine conditions is essential to ensure regular production and maintain equipment. Traditional approaches, such as reactive and preventive maintenance, often prove to be inefficient in terms of resource and time management. This study explores the application of predictive maintenance in companies, leveraging data science and machine learning techniques. Predictive maintenance anticipates failures by analyzing real-time collected data using advanced algorithms. This reduces unplanned downtimes, optimizes interventions, and improves resource management, thereby increasing equipment availability and operational performance. A comparison of machine learning algorithms, including decision trees, XGBoost, SVM, KNN, logistic regression, Gaussian Naive Bayes, and random forests, shows that XGBoost offers superior classification accuracy. Consequently, this model was chosen to develop a user-friendly application that allows users to easily monitor machine health. Keywords: Predictive maintenance, Industry 4.0, Diagnostics, Decision tree algorithms, Machine Learning, Internet of Things, Smart manufacturing, Logistic regression, Gaussian Naive Bayes, Random forest.

Keywords: Neural networks, Predictive control, Complex systems
Abstract: Ensuring optimal water quality in large-scale river systems remains a critical environmental challenge. Poor water quality can significantly affect aquatic ecosystems, leading to increased fish mortality, impaired health, and disruptions in biodiversity. While traditional methods like the Water Quality Index (WQI) and Water Quality Classification (WQC) are constrained by their reliance on periodic laboratory analyses, recent studies have shifted towards using sensor data combined with machine learning (ML) models for more accurate and real-time monitoring. This study enhances water quality monitoring by accurately predicting Dissolved Oxygen (DO) levels, which are a key indicator of water quality, in the rivers of Flanders, Belgium, through the application of machine learning models. The results showed that the Long Short-Term Memory (LSTM) model outperformed other models in capturing the intricate temporal patterns of dissolved oxygen (DO) variations. It demonstrated robust performance in both single-step and multi-step predictions, particularly in detecting critical DO levels (<6 mg/L), which are indicative of poor water quality. Additionally, the incorporation of confidence intervals into the predictions provided a more reliable assessment of forecasting performance. The findings of this study establish a robust predictive framework for large-scale water quality monitoring, providing valuable insights for the protection and mitigation of aquatic ecosystems.

Keywords: Intelligent control systems, Complex systems, Optimisation
Abstract: Additive Manufacturing (AM) is revolutionizing production by enabling complex designs with minimal waste and reduced production times. This paper introduces a Model-Based Systems Engineering (MBSE) approach to enhance AM processes through the Advisor System for Additive Manufacturing (ASAM). Utilizing the CESAM architecture and SysML, ASAM creates integrated multi-architecture models to improve design validation, flaw detection, and process optimization. The framework aims to include real-time optimization and predictive analytics in future versions. Initial results from aerospace, healthcare, and education demonstrate ASAM's effectiveness in standardizing AM workflows, suggesting significant potential for increasing efficiency and scalability. This version keeps the focus on the key innovations and intended future developments of your research while preserving the academic integrity and reference structure.

Keywords: Intelligent control systems, Neural networks, Linear systems
Abstract: As directional drilling technologies evolve, effective control systems are crucial for optimizing drilling trajectories in complex subsurface formations. This paper presents an advanced control strategy using Gated Recurrent Unit (GRU) neural networks to achieve real-time trajectory control in directional drilling operations. The proposed system integrates GRU-based adaptive learning with finite element modeling (FEM) to dynamically update the parameters of a PID controller. By continuously adjusting PID gains based on real-time feedback and error minimization, the system improves adaptability, robustness, and precision in downhole conditions. The GRU network efficiently captures temporal dependencies, enabling predictive control and minimizing trajectory deviations. In addition, real-time data feedback further improves control accuracy and operational efficiency. The simulation results illustrate the effectiveness of the GRU-based adaptive PID control approach, which demonstrates improved trajectory prediction and system stability in complex drilling environments.

Keywords: Intelligent control systems, Predictive control, Aerospace control
Abstract: Quadcopter attitude control is considered a challenging task, due to the inherent nonlinearities and various factors that affect system behavior. In this work, we develop a data-driven nonlinear model predictive control (MPC) framework that can successfully address these challenges. To achieve this, the proposed scheme incorporates radial basis function (RBF) neural network models for predicting the quadcopter orientation dynamics; in contrast to first principles-based predictive models, which are prone to modeling uncertainties, the data-driven nature of RBF models helps them to capture phenomena like aerodynamic effects, thus, leading to more accurate quadcopter control. The resulting controller is evaluated on a simulated quadcopter, while a comparison to a nonlinear MPC scheme using first principles-based predictive models validates the superiority of the proposed approach.

Keywords: Intelligent transportation systems, Automotive control
Abstract: Improving trust in the operation of intelligent transportation systems (ITSs) is an actual challenge for overcoming the trough of disillusionment in the development phase regarding autonomous vehicles (AVs). There are identified critical gaps in the field that motivate the development of new theoretically grounded methods: most of the existing methods can be used for specific systems without generality in applicability, and the achieved explainability level is aimed only engineers and specialists. This paper aims to provide a method for developing an explainable representation on a specific ITS, such as intersection management with AV. The challenge is to find a transformation method which the explainable representation is resulted in. In this paper a decision-tree-based solution is proposed that results in a low-order approximating system in explainable form. It is presented an optimization method that results in the decision tree through the selection of its parameters, focusing on the selected ITS problem. The achieved rules within the explainable representation are used for supporting the human driving strategy in order to reduce critical interactions between AVs and human-driven vehicles. The effectiveness of the method is illustrated through high number of simulation scenarios. The outcome of the simulation is that the number of critical and risky interactions can be significantly reduced, if the rules from the explainable representation are considered.

Keywords: Neural networks, Image processing, Marine control
Abstract: The persistent interaction between dolphins and commercial fishing operations has led to ecological and socio-economic challenges, primarily through bycatch and depredation. Traditional mitigation strategies have shown limited success, needing innovative solutions. Intelligent robotic systems capable of identifying and consequently responding to dolphin vocalizations seem to be a promising approach to mitigate dolphin interactions with fishing operations. The core of this intelligent system should be an advanced algorithm or an artificial intelligence architecture capable of identifying dolphin vocalizations and distinguishing them from other underwater sounds. Thus, this study proposes a novel approach to detect dolphin whistles using a convolutional neural network (CNN) paired with advanced spectrogram processing techniques. The method utilizes audio recordings of common bottlenose dolphins (Tursiops truncatus) from Oltremare marine park in Italy. Whistle detection was enhanced by applying edge-detection filters to spectrograms, which highlights characteristic of dolphin whistles while filtering out noise. The processed spectrograms served as inputs to a CNN with a three-layer architecture optimized for binary classification of dolphin whistles. The model achieved very promising results, with accuracy, precision, recall, and F1-scores around 99% across a 10-fold cross-validation. The findings demonstrate the method robustness, offering potential applications in conservation efforts and real-time monitoring. Future research will focus on adapting the approach to field conditions where real-time processing and non-ideal whistle recording pose additional challenges.

Keywords: Neural networks, Signal processing, Marine control
Abstract: The interaction between dolphins and fishing activities poses economic and ecological challenges, needing improved monitoring techniques. The present study presents a preliminary investigation into the classification of bottlenose dolphin (Tursiops truncatus) vocalizations using convolutional neural networks (CNNs), applied to a dataset of underwater acoustic recordings. The proposed approach classified four main vocalization typologies — whistles, echolocation clicks, burst pulse sounds, and feeding buzzes — while distinguishing them from background noise. The dataset, collected at the Oltremare marine park in Riccione, Italy, was processed through spectrogram analysis, with the application of specific filtering to enhance signal characteristics and to filter out the undesired noise. The CNN model, trained using a 10-fold cross-validation approach, achieved an average classification accuracy exceeding 95%, with precision, recall, and F1-score close to 90%. The results demonstrate the feasibility of deep learning-based classification but also highlight the need to work with wide and differentiated datasets, particularly for identifying feeding buzzes. This preliminary study may contribute to the development of an autonomous monitoring system for detecting dolphin presence in marine environments using AI-based classification. Future work will focus on expanding the dataset with recordings from varied environments and optimizing preprocessing techniques to improve robustness in real-world conditions.

Keywords: Time-delay systems, Disturbance rejection, Modelling and simulation
Abstract: Recent research has demonstrated that the Harmonic Control Array (HCA) approach provides an effective control strategy for systems with periodic references or disturbances. To achieve zero steady-state error, the harmonic correction algorithm modifies the complex levels of the system input's harmonic components appropriately. This paper presents compensation of multi-harmonic periodic disturbances in time-delay systems utilizing Harmonic Control Arrays (HCA). The study evaluates the efficacy of the technique using a first-order time-delay system subject to a multi-harmonic periodic disturbance. Key performance metrics include disturbance rejection, robustness, adaptability, and implementation complexity. Simulation results demonstrate that the approach can effectively attenuate disturbances and achieves perfect steady-state reference tracking for each harmonic considered. HCA also exhibits superior robustness to parameter variations and various perturbations, as well as adaptability to changes in disturbance characteristics. The findings suggest that HCA offers significant advantages in addressing multi-harmonic periodic disturbances in time-delay systems, particularly in scenarios requiring high performance and adaptability involving a high number of harmonics.

Keywords: Time-delay systems, Unmanned systems, Autonomous systems
Abstract: The remote control of automated vehicles is highly affected by communication time delays, that can compromise stability, control accuracy, and passenger comfort. This paper introduces a dynamic speed limitation function designed to mitigate the negative effects of communication time delays by dynamically adjusting the vehicle's speed while maintaining a minimum stability margin. The proposed approach is validated through simulations, demonstrating substantial improvements in lateral control performance and ride comfort in teleoperated driving scenarios.

Keywords: Process control, Industrial automation, manufacturing, Time-delay systems
Abstract: This paper presents a relay based procedure for autotuning Proportional-Integral-Derivative (PID) controllers based on a short relay test, aimed at improving the efficiency and accuracy of PID tuning. Unlike traditional approaches, the proposed autotuner requires a single, brief experimental test performed for the duration of 2 times the settling time of the process. The goal is to obtain the process frequency response value and it's derivative at a certain frequency of interest. Then, any method can be utilized to tune the PID parameters. A highly nonlinear process featuring stochastic disturbances is chosen as the experimental case study of the proposed approach, consisting of vertical take-off and landing platform. The obtained results showcase the effectiveness of the proposed method in handling complex, non-linear systems. Results demonstrate that the novel autotuning approach proves easy to implement, making it highly suitable for real-world industrial applications and is immune to stochastic disturbances.

Keywords: Decentralized control, Time-delay systems, Robust control
Abstract: Decentralized stabilizing controller design problem for a system to achieve robust tracking of certain reference signals despite certain disturbances and modeling uncertainties is considered for linear time-invariant descriptor-type distributed-time-delay systems. The necessary and sufficient conditions for the solvability of this problem are presented. The structure of the controller which solves this problem, when a solution exists, is also given.

Keywords: Linear systems, Algebraic and geometric methods, Feedback stabilization
Abstract: We present new polynomial-based methods for SISO discrete-time quaternionic systems, highlighting how noncommutative multiplication modifies classical control approaches. Defining quaternionic polynomials via a backward-shift operator, we examine left and right fraction representations of transfer functions, showing that right zeros correspond to similarity classes of quaternionic matrix right eigenvalues. We then propose a feedback design procedure that generalizes pole placement to quaternions—a first approach using a genuine quaternionic polynomial equation. An illustrative example demonstrates its effectiveness.

Keywords: Linear systems, Computational methods, Feedback stabilization
Abstract: In this paper, a new data-driven input allocation procedure is proposed for strongly input redundant systems, generalizing a recent data-based representation result for LTI systems. This algorithm assumes a Linear Parameter Varying (LPV) structure to implicitly model the unknown system using only collected input-state data, and directly computes the null space basis of the input matrix for a linear dynamic input allocator. A complete data-driven LPV control and allocation scheme are then presented implementing a recent semidefinite LPV control program for quadratic stabilization. Finally, this scheme is tested in simulation on a overactuated marine vessel system, along with an analysis of its practical usage and performance evaluation.

Keywords: Feedback stabilization, Disturbance rejection, Power systems and smart grid
Abstract: In this paper, a robust H_{infty} integral state-feedback controller, providing optimal stability as well as robust performance suitable for implementation in DC-DC converters, is proposed. Under disturbances arising from variations in input voltage and load, and uncertainties from converter's parameters, the suggested control approach is developed to obtain the gains of the integral state-feedback controller by means of linear matrix inequality constraints based on Lyapunov stability theory. The performance of the designed controller under various scenarios is presented using experimental setup. The obtained results affirm the efficacy of the suggested design methodology.

Keywords: Linear systems
Abstract: This paper presents a data-driven approach to simultaneously estimating the inputs and states with a delay of a linear discrete-time system. We consider both noise-free and noisy data cases. In the case of noise-free data, we develop an algorithm to reconstruct inputs and states with a delay simultaneously. We note that a system property known as system invertibility plays an important role in developing this algorithm. Furthermore, we prove that the algorithm returns uniquely the inputs and states of the system. Building on this, in the noisy data case, we present a recursive algorithm akin to the celebrated Kalman filter, which estimates the inputs and states with a delay simultaneously. We consider an example to illustrate the developed results.

Keywords: Feedback stabilization, Linear systems, Modelling and simulation
Abstract: This paper addresses the problem of control synthesis for a class of Two-Dimensional (2-D) singular continuous systems described by a Roesser model. By transforming the original system into an equivalent separated standard form and employing the Lyapunov function in combination with specific slack variables, sufficient Linear Matrix Inequality (LMI) conditions are derived. Consequently, a Static Output Feedback (SOF) controller is designed to stabilize an initially unstable system and guarantee its admissibility. The effectiveness of the proposed method is then demonstrated through a numerical example.

Keywords: Multi-agent systems, Modelling and simulation
Abstract: We propose a multi-agent model framework for crowd dynamics in pedestrian environments. The model integrates physical and psychological interactions among agents and with the environment, accounting for physical and social forces based on inter-agent and obstacle distances. It improves the physical characteristics of an agent by modeling the agent's active and reactive torso rotation. These features capture an agent's rotational and lateral movement induced by torque interactions between the agents and enable phenomena such as shouldering and navigating narrow corridors. The physical layer also includes a feedback control force, specifically Proportional-Integral (PI) feedback control, that enables the agent to track a desired velocity profile, reflecting directional intention and determination. The authority of this control force is determined by using upper limits on the allowable force magnitude, which is determined by the competitive index characteristic of each agent. The psychological layer models agent competitiveness, incorporating inter-agent interactions and responses to environmental hazards. The physical and psychological layers are coupled by the inter-agent distances and the psychological state, determining the feedback control authority on each agent’s pushing force and motion. To illustrate the novel behavior enabled by this model, we present extensive simulation scenarios highlighting the model's layers and how the different parameters influence the crowd's behavior.

Keywords: Modelling and simulation, Intelligent control systems, marine
Abstract: Feature selection (FS) is a critical step in ma- chine learning (ML) applications, particularly in the maritime transportation domain, where large-scale data from sensors, weather conditions, and operational logs can introduce re- dundancy and noise. Feature selection plays a crucial role in improving ML model performance by reducing dimen- sionality, enhancing generalization, and mitigating overfitting. However, its impact extends beyond predictive accuracy to the explainability of AI (XAI), influencing how interpretable and transparent models become. In this study, we systematically compare various feature selection methods—including filter, wrapper, and embedded techniques—analyzing their effects on model performance and explainability. We assess multiple ML models across different datasets to evaluate trade-offs between predictive accuracy, computational efficiency, and ex- plainability. The study leverages shapley additive explanations to compute feature importance scores to quantify how feature selection impacts model transparency. Our findings highlight that while some FS methods enhance predictive accuracy, they may compromise interpretability, whereas others strike a balance between accuracy and explainability.

Keywords: Disturbance rejection, Aerospace control, Modelling and simulation
Abstract: Aerial vehicles employed for inspection activities require high resilience against environmental disturbances to maintain a stable observation state. In order to effectively counteract those disturbances, the allocation of direct forces, particularly in the medial and lateral directions, gains importance. This raises the necessity of including those two directions in the controllability analysis of the vehicle. Hence, this work extends a methodology for calculating the Attainable and Required Reaction Sets from four to six dimensions, adding the medial and lateral directions to the previously considered vertical, pitch, roll, and yaw dimension. The Attainable Reaction Set (ARS) represents the forces and moments that the aerial vehicle is capable of generating. This set is evaluated against the Required Reaction Set (RRS), which establishes the required forces and moments defined by the mission. The six-dimensional methodology is applied to a tilted-propeller multicopter designed to operate in marine conditions with high environmental disturbances. As a result, the required and attainable lateral and medial forces, along with the other four dimensions, can be clearly analyzed. The four-dimensional algorithm is applied to the same model and allows the six-dimensional methodology to be validated and contrasted in the common dimensions.

Keywords: Fuzzy logic and fuzzy control, Marine control, Modelling and simulation
Abstract: This paper investigates the modeling and control design problem for common rail injection systems in Diesel marine engines, with a focus on H_{infty} static output feedback (SOF) control. The disturbance signal is assumed to have a known frequency range, constrained within a finite frequency (FF) domain. The primary goal is to maintain the pump pressure and rail pressure at their desired reference levels by adjusting the engine speed and the injector actuation signals. To accurately capture the system's nonlinear dynamics, a model based on equilibrium points is employed. Subsequently, an SOF controller is synthesized to achieve robust pressure regulation while guaranteeing H_{infty} performance, expressed as a set of linear matrix inequalities (LMIs). Numerical simulations are carried out to validate the effectiveness and robustness of the proposed control strategy.

Keywords: Computational intelligence, Modelling and simulation
Abstract: Power prediction is crucial for dynamic ship routing, a key strategy for reducing fuel consumption and minimizing environmental impact. Ships, equipped with some of the largest and most powerful engines, contribute significantly to global pollution. Accurate power prediction enables naval architects to evaluate different routes and select the most fuel-efficient option. This study presents a comprehensive comparison of three power prediction methods, assessing their performance for maritime applications. The first method is a physics-based approach using computational fluid dynamics data to estimate power requirements. The second method is a data-driven model, relying on historical data inputs such as speed, draft, and wind speed. The third method is a hybrid approach that integrates computational fluid dynamics data interpolations with machine learning techniques. Results highlight the strengths and limitations of each method, offering valuable insights for optimizing ship routing and power management strategies.

Keywords: Decentralized control, Distributed systems, Multi-agent systems
Abstract: In this paper we present a decentralized control framework for swarm density regulation using Partial Differential Equation (PDE)-constrained optimization, implemented with a Model Predictive Control (MPC) paradigm. We propose a novel formulation which allows each agent in the swarm to optimize its control input using locally available information only. Extensive numerical experiments show the effectiveness of the developed decentralized control framework and compare its performance to centralized control strategies. Real robot experiments validate the proposed algorithm.

Keywords: Cyber-physical systems, Networked systems
Abstract: This paper examines the detection of replay attack in a control system that is in steady state at some operating point. A watermarking scheme based on switching multi-sine waves is proposed in which the controller adds the watermark to the control signal and monitors the corresponding effects on sensor readings received through a communication channel. The frequencies of sinusoids are changed frequently to prevent the attacker from replicating the effects of watermarking in the system output. Switching sinusoids could generate undesirable transient responses. This paper presents a procedure to choose an appropriate number of sinusoidal terms, and relative amplitudes and phases for the sine waves, to prevent such undesirable transients. The presence (or absence) of watermarking signal in the output is examined using power spectral density estimation.

Keywords: Distributed systems, Networked systems, Multi-agent systems
Abstract: In a rapidly evolving industrial environment, securing communications is a fundamental need. In particular, safeguarding the integrity and confidentiality of information exchanged within the industrial network remains a critical priority. For this purpose, a two-level architecture has been developed that includes: (i) the distributed network layer that implements an average gossip protocol while applying a geometric encryption strategy aimed at preserving confidentiality; (ii) a higher-level verifier that is able to detect false data injection attacks and remediate them through a compensation mechanism. Thanks to the compensation strategy performed by the verifier, which communicates ad-hoc quantities to the benign agents, it will be possible for them to converge to the average of all initial conditions, including those of the nodes flagged as malicious. The paper is correlated with a simulation campaign aimed at establishing the effectiveness of the theoretical results.

Keywords: Education and training, Robotics
Abstract: The introduction of control theory in education, beginning in primary school, is an essential step toward preparing students for 21st-century challenges. Control theory’s inclusion in STEAM education is critical as a foundational element of technological and engineering advancements. Teaching these concepts to young learners requires innovative methods that balance accessibility, engagement, and adaptability. This study integrates an AI-enhanced chatbot to support Educational Robotics activities with the potential to create digital twins of learning classrooms. A total of 39 primary school students participated in the first experimental session. Questionnaire responses revealed high levels of engagement and interest, with the majority of students expressing enthusiasm for the chatbot as a companion. Finally, by leveraging data from AI-student interactions, the proposed infrastructure can model individual learning paths and classroom dynamics, enabling precise interventions and dynamic insights. This capability holds the potential not only to enhance educational outcomes but also to support teachers by providing real-time feedback on student progress, enabling them to adapt their instructional strategies effectively, and providing a scalable framework for advancing control education and STEAM learning.

Keywords: Nonlinear systems, Fuzzy logic and fuzzy control, Event based systems
Abstract: This paper presents a novel dynamic-event-triggered output-feedback control strategy for stabilizing nonlinear systems represented by polynomial fuzzy models. By incorporating an event-triggering mechanism, the proposed controller significantly reduces unnecessary control updates, thereby lowering computational and communication burdens compared to conventional sampled-data output-feedback controllers. The stability of the closed-loop system is analyzed using a Lyapunov–Krasovskii functional in combination with a sum-of-squares (SOS) optimization framework. The resulting stability conditions ensure asymptotic stability while optimizing resource utilization. Simulation results validate the effectiveness of the proposed method, demonstrating a substantial reduction in control transmissions without sacrificing system stability or performance.

Keywords: Networked systems, Energy efficient systems, Decentralized control
Abstract: This paper explores the application of Hierarchical Reinforcement Learning (HRL) in optimizing electric vehicle (EV) charging, addressing challenges in load scheduling, energy cost management, and real-time dynamic pricing. We propose a hierarchical environment framework with two interconnected levels designed to efficiently manage charging demands across multiple Charging Stations (Chargym). The upper level focuses on dynamic pricing optimization, while the lower level handles load distribution among EVs. The Deep Deterministic Policy Gradient (DDPG) algorithm is implemented within this framework and evaluated against a baseline to assess its performance. Experimental results demonstrate that HRL effectively decomposes the complex EV charging problem into manageable subtasks, achieving improved efficiency in scheduling and pricing while ensuring cost-effective energy distribution. The upper-level DDPG agent in our formulation has shown a 35.58% improvement, and the overall DSO profit has shown a 101.04% improvement, when compared to the RBC baseline.

Keywords: Optimisation, Adaptive control, Process control
Abstract: This paper presents a Bayesian Optimization (BO)-based method for online tuning of a Proportional-Integral- Derivative (PID) controller in a real-world thermal zone HVAC application. Unlike conventional PID tuning methods that require manual intervention and offline calibration, the proposed approach optimizes PID gains autonomously and continuously, adapting daily to external disturbances (e.g., weather fluctuations, occupancy changes) without requiring a system model. A novel transition error mechanism accounts for occupancy patterns, enabling the controller to anticipate heating/cooling demands by preemptively adjusting set-points, thereby improving thermal comfort and energy efficiency. Through a 90-day real-world experiment under varying winter conditions (January–March), the BO-optimized PID controller is evaluated against a traditional on-off dead-band rule-based controller (RBC), demonstrating 5–7% energy savings despite lacking direct energy consumption feedback during tuning, a 45% reduction in temperature deviations from set-points, and faster convergence to target temperatures during occupancy transitions. These findings validate BO as a robust, model-free solution for real-time PID tuning in dynamic environments, balancing energy efficiency with occupant preferences.

Keywords: Optimisation, Decentralized control, Biomedical engineering
Abstract: Errors are most likely to occur during the maintenance phase of anesthesia, which not only ensures sustained sedation, but also that a patient is kept within normal parameters. Making use of an existing model for this phase, we explore in this paper the efficacy of multi-objective genetic algorithms (MOGA) in optimizing the parameters of a pair of proportional integrative derivative controllers (PID) within a decentralized configuration. The algorithm manages to reach varied solutions, that successfully meet most of the imposed criteria.

Keywords: Optimisation, Neural networks, Nonlinear systems
Abstract: Region of Attraction (ROA) estimation with accuracy is crucial for effective control design in nonlinear dynamical systems. Sum-of-squares (SOS) optimization refines Lyapunov function representations and expands ROA estimation for polynomial dynamical systems. However, traditional SOS methods tend to be overly conservative. In contrast, deep learning has emerged as a powerful tool in robotics, enabling data-driven ROA estimation. While deep learning offers strong empirical performance, its lack of stability guarantees remains challenging in safety-critical applications. This paper compares two ROA estimation methods: Sum-of-squares (SOS) optimization and Neural Network Lyapunov Functions (NLFs). We examine their effectiveness in approximating Lyapunov functions for stability assessment, highlighting their strengths, limitations, and practical relevance. Using the Van der Pol oscillator as a benchmark, we conduct detailed simulations to evaluate each method’s performance. We provide a comprehensive comparison, considering computational efficiency, accuracy, and scalability, offering insights into their applicability in real-world aerospace systems.

Keywords: Optimisation, Renewable energy and sustainability, Energy efficient systems
Abstract: This paper presents a model predictive control (MPC) algorithm for optimizing the energy operations of a smart beehive equipped with a photovoltaic panel, an electrical storage, and a vision system for automatic detection of Varroa-infested bees. A camera at the entrance of the beehive monitors the entering and exiting bees, and image processing algorithms are used to detect the presence of Varroa. The goal of the proposed energy management algorithm is to maximize the number of monitored bees (image acquisition and processing consume energy), while ensuring that the beehive does not run out of energy, so that monitoring can be ensured over a prolonged time. Simulation results on real data show that the propose controller effectively manages the beehive, ensuring the achievement of the above objectives.

Keywords: Optimisation, Robotics, Modelling and simulation
Abstract: Automated mushroom harvesting is challenging, especially in vertically stacked growing environments, due to limited space, the delicate nature of the crop, and the need for precise robot motion. This paper presents CeuiBot, a robotic system composed of two SCARA-type manipulators, each mounted on a linear rail, and installed on a mobile base that moves between shelves. The system is designed to operate efficiently in constrained agricultural environments. Unlike traditional approaches that focus on avoiding joint limits or minimizing torque, our contribution is at the control level. We propose an optimization strategy that reduces redundant motion by minimizing prismatic rail displacement and limiting the simultaneous activation of multiple joints. Our method is based on a workspace-aware inverse kinematics formulation that prioritizes minimal rail movement while maintaining target reachability. A workspace segmentation technique is also introduced to avoid singularities and self-collisions. Experimental results show a significant improvement in trajectory efficiency, reduced actuator usage, and optimized energy consumption. The proposed approach is efficient, robust, and compatible with real-time computation, making it suitable for selective mushroom harvesting in confined farming environments.

Keywords: Hybrid systems, Predictive control, Optimisation
Abstract: The increasing integration of electric vehicles (EVs) into power systems presents both challenges and opportunities for energy management. In this paper, we propose a two-level hierarchical model predictive control (MPC) framework for optimizing energy transactions in bidirectional charging hubs, focusing on load shifting as a long-term demand response (DR) strategy. At the higher level, an optimization problem based on portfolio optimization is solved every 15 minutes to determine the optimal energy packets allocated to EVs, the energy buffer, and grid interactions, considering energy prices and grid demands. This problem can be relaxed to a formulation as a mixed-integer linear program (MILP). At the lower level, a quadratic optimization problem with linear constraints determines the charging and discharging power of each EV with a higher time resolution to ensure that the assigned energy packets are accurately delivered while adhering to operational constraints. The proposed approach enables scalable and flexible load shifting, allowing bidirectional charging hubs to manage large-scale EV fleets while contributing to DR programs efficiently. Unlike traditional short-term optimization methods, our framework considers a long-term planning horizon, exploiting the flexibility provided by EVs and energy buffers to optimize energy trading decisions dynamically. Simulation results demonstrate the effectiveness of the proposed hierarchical control strategy.