Pub Date : 2025-09-16DOI: 10.1016/j.ifacsc.2025.100339
Teketel Ketema , Surafel Luleseged Tilahun , Simon D. Zawka , Abebe Geletu
This paper proposes a deep Koopman-based reachability analysis technique for a data-driven control of unknown nonlinear systems subject to process and measurement noises. An intelligent approach combining a neural network and Q-learning algorithm is employed. In particular, the power of the Long Short-Term Memory (LSTM) neural network is leveraged to lift the original nonlinear system into a higher-dimensional space, where the nonlinear dynamics can be approximated linearly, relying solely on the input–output data. The LSTM is set to draw learning insights from Extended Dynamic Mode Decomposition (EDMD) and Information-Theoretic Metric Function (ITMF) results. The Q-learning algorithm is employed to compute adaptive input–output references in the implementation of an adaptive nonlinear zonotopic predictive control technique to compute a robust control input of the system. We also introduced controllability and observability criteria in the presence of noisy data. Finally, a numerical example is given to verify the proposed approach.
{"title":"Deep Koopman-based reachability analysis for data-driven predictive control of unknown nonlinear systems","authors":"Teketel Ketema , Surafel Luleseged Tilahun , Simon D. Zawka , Abebe Geletu","doi":"10.1016/j.ifacsc.2025.100339","DOIUrl":"10.1016/j.ifacsc.2025.100339","url":null,"abstract":"<div><div>This paper proposes a deep Koopman-based reachability analysis technique for a data-driven control of unknown nonlinear systems subject to process and measurement noises. An intelligent approach combining a neural network and Q-learning algorithm is employed. In particular, the power of the Long Short-Term Memory (LSTM) neural network is leveraged to lift the original nonlinear system into a higher-dimensional space, where the nonlinear dynamics can be approximated linearly, relying solely on the input–output data. The LSTM is set to draw learning insights from Extended Dynamic Mode Decomposition (EDMD) and Information-Theoretic Metric Function (ITMF) results. The Q-learning algorithm is employed to compute adaptive input–output references in the implementation of an adaptive nonlinear zonotopic predictive control technique to compute a robust control input of the system. We also introduced controllability and observability criteria in the presence of noisy data. Finally, a numerical example is given to verify the proposed approach.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"34 ","pages":"Article 100339"},"PeriodicalIF":1.8,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-09DOI: 10.1016/j.ifacsc.2025.100338
Khadija El Harouri , Soumia El Hani , Nisrine Naseri , Imade Aboudrar , Amina Daghouri
Electric vehicles (EVs) are becoming a basis of sustainable mobility, requiring efficient charging management to minimize costs, balance grid demand, and optimize renewable energy utilization. In workplace parking lots, integrating solar energy and vehicle-to-grid (V2G) technology offers new opportunities for smart energy management. This paper presents an optimization-based charging strategy using Particle Swarm Optimization (PSO) to minimize total energy costs while reducing peak power drawn from the grid, maximizing the use of photovoltaic (PV) energy and ensure that all vehicles reach their target State of Charge (SOC) before leaving the parking lot. Additionally, The proposed approach leverages advantage of V2G technology, enabling EVs to return energy to the grid during peak demand hours, which enhances grid stability and reducing overall energy expenses. A key contribution of this work is the comparative analysis of EV charging management in three different geographical contexts: Morocco, France, and Tunisia. Each country provides distinct energy cost structures, solar availability. A dynamic electricity pricing model is incorporated to adapt the charging strategy based on daily and seasonal tariff variations. The optimization strategy considers multiple constraints like EV arriving and leaving periods, initial and target SOC, PV energy production, and dynamic electricity pricing. Results from simulations indicate that the suggested PSO-based charging strategy achieves significant cost savings can reach up to 65% compared to a conventional unmanaged scenario, reduces peak power coming from the grid, and maximize PV power utilization via self-consumption. Additionally, the findings highlight the benefits of multi-objective optimization in smart parking energy management.
{"title":"Optimizing electric vehicle charging in smart parking lots using particle swarm optimization: A comparative study in Morocco, France, and Tunisia","authors":"Khadija El Harouri , Soumia El Hani , Nisrine Naseri , Imade Aboudrar , Amina Daghouri","doi":"10.1016/j.ifacsc.2025.100338","DOIUrl":"10.1016/j.ifacsc.2025.100338","url":null,"abstract":"<div><div>Electric vehicles (EVs) are becoming a basis of sustainable mobility, requiring efficient charging management to minimize costs, balance grid demand, and optimize renewable energy utilization. In workplace parking lots, integrating solar energy and vehicle-to-grid (V2G) technology offers new opportunities for smart energy management. This paper presents an optimization-based charging strategy using Particle Swarm Optimization (PSO) to minimize total energy costs while reducing peak power drawn from the grid, maximizing the use of photovoltaic (PV) energy and ensure that all vehicles reach their target State of Charge (SOC) before leaving the parking lot. Additionally, The proposed approach leverages advantage of V2G technology, enabling EVs to return energy to the grid during peak demand hours, which enhances grid stability and reducing overall energy expenses. A key contribution of this work is the comparative analysis of EV charging management in three different geographical contexts: Morocco, France, and Tunisia. Each country provides distinct energy cost structures, solar availability. A dynamic electricity pricing model is incorporated to adapt the charging strategy based on daily and seasonal tariff variations. The optimization strategy considers multiple constraints like EV arriving and leaving periods, initial and target SOC, PV energy production, and dynamic electricity pricing. Results from simulations indicate that the suggested PSO-based charging strategy achieves significant cost savings can reach up to 65% compared to a conventional unmanaged scenario, reduces peak power coming from the grid, and maximize PV power utilization via self-consumption. Additionally, the findings highlight the benefits of multi-objective optimization in smart parking energy management.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"34 ","pages":"Article 100338"},"PeriodicalIF":1.8,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-09DOI: 10.1016/j.ifacsc.2025.100336
Birthe Göbel , Alexander Richter , Stefan J. Rupitsch , Alexander Reiterer , Knut Möller
Minimally invasive laparoscopic surgery, where endoscopes and instruments are inserted through small incisions, has advanced to the next stage of development: robot-assisted minimally invasive surgery. These systems use a camera and robotic manipulators operated by human surgeons through human-in-the-loop control. To further improve surgical precision and autonomy, data-driven assistance must be expanded. One promising approach is 3D reconstruction based on endoscopic images. A 30°endoscope tip enhances the field of view by enabling rotational motion around the instrument’s axis. However, when performing a 3D scan with such an endoscope, a blind spot inherently forms along the shaft axis, creating a region that cannot be captured during rotation. Additional missing data may arise due to occlusions from anatomical geometry and the specific endoscope pose during a scan. These limitations result in incomplete 3D reconstructions, which can negatively impact surgical navigation and decision-making. This paper presents a method tailored to medical applications for detecting and characterizing holes in laparoscopic 3D scans. The proposed method uses geometric analysis of the point cloud to identify regions of sparse or missing data and correlates these gaps with endoscope positioning and anatomical visibility. It is designed to operate robustly on high-density point clouds generated by advanced laparoscopic 3D reconstruction systems. By integrating robotic control, our method provides a foundation for adaptive endoscope repositioning to recover missing views and improve reconstruction completeness. The proposed method paves the way towards fast (5 s) feedback for optimized 3D scanning in laparoscopic environments.
{"title":"3D-Scan hole detection for robot-assisted laparoscopic surgery","authors":"Birthe Göbel , Alexander Richter , Stefan J. Rupitsch , Alexander Reiterer , Knut Möller","doi":"10.1016/j.ifacsc.2025.100336","DOIUrl":"10.1016/j.ifacsc.2025.100336","url":null,"abstract":"<div><div>Minimally invasive laparoscopic surgery, where endoscopes and instruments are inserted through small incisions, has advanced to the next stage of development: robot-assisted minimally invasive surgery. These systems use a camera and robotic manipulators operated by human surgeons through human-in-the-loop control. To further improve surgical precision and autonomy, data-driven assistance must be expanded. One promising approach is 3D reconstruction based on endoscopic images. A 30°endoscope tip enhances the field of view by enabling rotational motion around the instrument’s axis. However, when performing a 3D scan with such an endoscope, a blind spot inherently forms along the shaft axis, creating a region that cannot be captured during rotation. Additional missing data may arise due to occlusions from anatomical geometry and the specific endoscope pose during a scan. These limitations result in incomplete 3D reconstructions, which can negatively impact surgical navigation and decision-making. This paper presents a method tailored to medical applications for detecting and characterizing holes in laparoscopic 3D scans. The proposed method uses geometric analysis of the point cloud to identify regions of sparse or missing data and correlates these gaps with endoscope positioning and anatomical visibility. It is designed to operate robustly on high-density point clouds generated by advanced laparoscopic 3D reconstruction systems. By integrating robotic control, our method provides a foundation for adaptive endoscope repositioning to recover missing views and improve reconstruction completeness. The proposed method paves the way towards fast (<span><math><mo>∼</mo></math></span>5 s) feedback for optimized 3D scanning in laparoscopic environments.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"34 ","pages":"Article 100336"},"PeriodicalIF":1.8,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-04DOI: 10.1016/j.ifacsc.2025.100337
Dana Zimmermann, Hans-Michael Kaltenbach
Exercise is an important component for glucose management in type 1 diabetes, but remains challenging for automated insulin delivery systems as altered glucose dynamics are difficult to model explicitly. Glucose monitoring data might enable data-driven approaches for learning these dynamics implicitly. We propose combining model predictive control with a reinforcement learning component to adjust basal insulin infusion rates for exercise. We train our model on a variety of exercise scenarios and demonstrate improved glucose control using two different frameworks. We evaluate how generalizable both frameworks are by personalizing a trained model with a small number of additional individual-specific training episodes.
{"title":"Reducing exercise-related hypoglycemia in automated insulin delivery with reinforcement learning","authors":"Dana Zimmermann, Hans-Michael Kaltenbach","doi":"10.1016/j.ifacsc.2025.100337","DOIUrl":"10.1016/j.ifacsc.2025.100337","url":null,"abstract":"<div><div>Exercise is an important component for glucose management in type 1 diabetes, but remains challenging for automated insulin delivery systems as altered glucose dynamics are difficult to model explicitly. Glucose monitoring data might enable data-driven approaches for learning these dynamics implicitly. We propose combining model predictive control with a reinforcement learning component to adjust basal insulin infusion rates for exercise. We train our model on a variety of exercise scenarios and demonstrate improved glucose control using two different frameworks. We evaluate how generalizable both frameworks are by personalizing a trained model with a small number of additional individual-specific training episodes.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"34 ","pages":"Article 100337"},"PeriodicalIF":1.8,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145027860","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-03DOI: 10.1016/j.ifacsc.2025.100335
Guilherme C. Duran, Edson K. Ueda, André K. Sato, Thiago C. Martins, Marcos S.G. Tsuzuki
Electrode displacement is a common source of error in Electrical Impedance Tomography (EIT), particularly in long-term or dynamic measurements where stable electrode contact is difficult to maintain. This study proposes a comprehensive machine learning framework to detect, classify, and correct electrode displacements prior to image reconstruction. The approach combines tree-based classifiers—such as XGBoost and LightGBM—and Convolutional Neural Networks (CNNs) to identify both the presence and location of displaced electrodes. These models were evaluated across a series of classification tasks with increasing complexity, demonstrating high accuracy even in scenarios involving multiple simultaneous displacements with different magnitudes. For the rectification of distorted voltage measurements, several Denoising Autoencoder (DAE) configurations were investigated. Electrode-specific DAEs trained on all displacement magnitudes achieved an average Mean Squared Error (MSE) reduction of 84.5%, while shift-based DAEs offered computational efficiency for coarse corrections. A hybrid strategy combining fast shift-based and high-accuracy electrode-specific models proved effective and scalable. The pipeline was validated using both synthetic datasets and real EIT measurements, confirming its robustness and generalization capabilities. These results support the use of learning-based correction schemes to improve the reliability and consistency of EIT in practical applications affected by electrode movement.
{"title":"Advancements in Electrical Impedance Tomography: Addressing electrode displacement with artificial neural networks","authors":"Guilherme C. Duran, Edson K. Ueda, André K. Sato, Thiago C. Martins, Marcos S.G. Tsuzuki","doi":"10.1016/j.ifacsc.2025.100335","DOIUrl":"10.1016/j.ifacsc.2025.100335","url":null,"abstract":"<div><div>Electrode displacement is a common source of error in Electrical Impedance Tomography (EIT), particularly in long-term or dynamic measurements where stable electrode contact is difficult to maintain. This study proposes a comprehensive machine learning framework to detect, classify, and correct electrode displacements prior to image reconstruction. The approach combines tree-based classifiers—such as XGBoost and LightGBM—and Convolutional Neural Networks (CNNs) to identify both the presence and location of displaced electrodes. These models were evaluated across a series of classification tasks with increasing complexity, demonstrating high accuracy even in scenarios involving multiple simultaneous displacements with different magnitudes. For the rectification of distorted voltage measurements, several Denoising Autoencoder (DAE) configurations were investigated. Electrode-specific DAEs trained on all displacement magnitudes achieved an average Mean Squared Error (MSE) reduction of 84.5%, while shift-based DAEs offered computational efficiency for coarse corrections. A hybrid strategy combining fast shift-based and high-accuracy electrode-specific models proved effective and scalable. The pipeline was validated using both synthetic datasets and real EIT measurements, confirming its robustness and generalization capabilities. These results support the use of learning-based correction schemes to improve the reliability and consistency of EIT in practical applications affected by electrode movement.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"34 ","pages":"Article 100335"},"PeriodicalIF":1.8,"publicationDate":"2025-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145061274","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-01DOI: 10.1016/j.ifacsc.2025.100333
Jordan F. Hill, Samuel Jackson, Mia Uluilelata, Samrath Sood, Jaimey A. Clifton, Ella F.S. Guy, J. Geoffrey Chase
Respiratory diseases affect 14% of New Zealand’s population, with over 100,000 individuals suffering from sleep apnoea, which causes breathing disruptions due to airway blockages. Positive airway pressure (PAP) devices are the gold standard treatment, typically operating in continuous (CPAP), bilevel (BiPAP), or automatic (APAP) modes. However, current PAP devices cost between NZ$800–$2500, creating a financial barrier for users, particularly those from lower socio-economic backgrounds. The mePAP was developed as a low-cost, open-source PAP device, constructed for NZ$250, capable of delivering CPAP, BiPAP, and APAP therapies with an airway pressure sensor for more precise control. Validation against a Fisher & Paykel CPAP was performed through benchtop testing, mechanical lung simulations, and a clinical trial with 40 healthy subjects. The mePAP was preferred by 42.5% of subjects, with 25% reporting no difference between the devices. A mean comfort rating of 6.36 for the mePAP compared to 5.92 for the Fisher & Paykel CPAP confirmed the two devices were comparable, with pressure fluctuations from the mePAP’s low-cost motor imperceptible to users. The airway sensor feedback loop enabled accurate pressure delivery, with BiPAP and APAP algorithms dynamically adjusting therapy pressure in response to breathing patterns. These results validate the mePAP as a low-cost alternative to commercial PAP devices, with comparable performance and comfort. Its affordability and open-source design have the potential to improve healthcare accessibility and reduce inequities, making respiratory therapy more accessible to underserved populations while enabling further research into respiratory treatments.
{"title":"Validation of the mePAP: A low-cost, high-quality, open-source PAP device for research and increasing equity in respiratory care","authors":"Jordan F. Hill, Samuel Jackson, Mia Uluilelata, Samrath Sood, Jaimey A. Clifton, Ella F.S. Guy, J. Geoffrey Chase","doi":"10.1016/j.ifacsc.2025.100333","DOIUrl":"10.1016/j.ifacsc.2025.100333","url":null,"abstract":"<div><div>Respiratory diseases affect 14% of New Zealand’s population, with over 100,000 individuals suffering from sleep apnoea, which causes breathing disruptions due to airway blockages. Positive airway pressure (PAP) devices are the gold standard treatment, typically operating in continuous (CPAP), bilevel (BiPAP), or automatic (APAP) modes. However, current PAP devices cost between NZ$800–$2500, creating a financial barrier for users, particularly those from lower socio-economic backgrounds. The mePAP was developed as a low-cost, open-source PAP device, constructed for NZ$250, capable of delivering CPAP, BiPAP, and APAP therapies with an airway pressure sensor for more precise control. Validation against a Fisher & Paykel CPAP was performed through benchtop testing, mechanical lung simulations, and a clinical trial with 40 healthy subjects. The mePAP was preferred by 42.5% of subjects, with 25% reporting no difference between the devices. A mean comfort rating of 6.36 for the mePAP compared to 5.92 for the Fisher & Paykel CPAP confirmed the two devices were comparable, with pressure fluctuations from the mePAP’s low-cost motor imperceptible to users. The airway sensor feedback loop enabled accurate pressure delivery, with BiPAP and APAP algorithms dynamically adjusting therapy pressure in response to breathing patterns. These results validate the mePAP as a low-cost alternative to commercial PAP devices, with comparable performance and comfort. Its affordability and open-source design have the potential to improve healthcare accessibility and reduce inequities, making respiratory therapy more accessible to underserved populations while enabling further research into respiratory treatments.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"33 ","pages":"Article 100333"},"PeriodicalIF":1.8,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144922557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-21DOI: 10.1016/j.ifacsc.2025.100334
Christopher Yew Shuen Ang , Yeong Shiong Chiew , Xin Wang , Ean Hin Ooi , Mohd Basri Mat Nor , Matthew E. Cove , Cong Zhou , J. Geoffrey Chase
Background and Objective
: Computerised decision support systems (CDSS) in mechanical ventilation (MV) provide individualised, closed-loop treatment but often require extensive input parameters, which are challenging to obtain continuously in clinical settings. Many also fail to incorporate mechanical power (MP) and MP ratio — recently identified as significant predictors of patient outcomes. This study introduces the Stochastic Virtual Patient Ventilation Protocol (SVP VENT), a model-based CDSS addressing these limitations.
Methods
: The SVP VENT Protocol integrates a stochastic virtual patient model to predict temporal lung elastance, , trends and deliver closed-loop, lung protective ventilation minimising MP ratio and driving pressure. The protocol was validated against the VENT and SiVENT protocols using an established virtual patient platform comprising over 1229 h of both volume control (VC) and pressure control (PC) retrospective MV data. Patient responses were monitored to ensure adherence to accepted clinical safety guidelines.
Results
: The SVP VENT protocol consistently outperformed retrospective clinical data, VENT and SiVENT protocols in ensuring adherence to clinical safety metrics, achieving an all-adherence rate of 57% and 67% for the VC and PC cohorts, respectively. Across cohorts, the protocol maintained MP and MP ratio levels below safety thresholds (12 J/min and 4.5, respectively), and extended intervention intervals up to 3 h, potentially reducing clinical workload.
Conclusion
: Overall, the virtual trial demonstrates the SVP VENT protocol’s potential to enhance MV management by extending intervention intervals, while maintaining patient safety. These findings support initial clinical trials to evaluate the protocol’s impact on clinical workload and patient safety over prolonged monitoring periods, facilitating its integration into standard clinical practices.
{"title":"Stochastic virtual patient-guided mechanical ventilation treatment: A virtual patient study with mechanical power consideration","authors":"Christopher Yew Shuen Ang , Yeong Shiong Chiew , Xin Wang , Ean Hin Ooi , Mohd Basri Mat Nor , Matthew E. Cove , Cong Zhou , J. Geoffrey Chase","doi":"10.1016/j.ifacsc.2025.100334","DOIUrl":"10.1016/j.ifacsc.2025.100334","url":null,"abstract":"<div><h3>Background and Objective</h3><div>: Computerised decision support systems (CDSS) in mechanical ventilation (MV) provide individualised, closed-loop treatment but often require extensive input parameters, which are challenging to obtain continuously in clinical settings. Many also fail to incorporate mechanical power (<em>MP</em>) and <em>MP ratio</em> — recently identified as significant predictors of patient outcomes. This study introduces the Stochastic Virtual Patient Ventilation Protocol (SVP VENT), a model-based CDSS addressing these limitations.</div></div><div><h3>Methods</h3><div>: The SVP VENT Protocol integrates a stochastic virtual patient model to predict temporal lung elastance, <span><math><msub><mrow><mi>E</mi></mrow><mrow><mi>r</mi><mi>s</mi></mrow></msub></math></span>, trends and deliver closed-loop, lung protective ventilation minimising <em>MP ratio</em> and driving pressure. The protocol was validated against the VENT and SiVENT protocols using an established virtual patient platform comprising over 1229 h of both volume control (VC) and pressure control (PC) retrospective MV data. Patient responses were monitored to ensure adherence to accepted clinical safety guidelines.</div></div><div><h3>Results</h3><div>: The SVP VENT protocol consistently outperformed retrospective clinical data, VENT and SiVENT protocols in ensuring adherence to clinical safety metrics, achieving an all-adherence rate of <span><math><mo>∼</mo></math></span>57% and <span><math><mo>∼</mo></math></span>67% for the VC and PC cohorts, respectively. Across cohorts, the protocol maintained <em>MP</em> and <em>MP ratio</em> levels below safety thresholds (12 J/min and 4.5, respectively), and extended intervention intervals up to 3 h, potentially reducing clinical workload.</div></div><div><h3>Conclusion</h3><div>: Overall, the virtual trial demonstrates the SVP VENT protocol’s potential to enhance MV management by extending intervention intervals, while maintaining patient safety. These findings support initial clinical trials to evaluate the protocol’s impact on clinical workload and patient safety over prolonged monitoring periods, facilitating its integration into standard clinical practices.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"33 ","pages":"Article 100334"},"PeriodicalIF":1.8,"publicationDate":"2025-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144917878","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-20DOI: 10.1016/j.ifacsc.2025.100332
Yuwei Sun, Samantha Couper, Cong Zhou, J. Geoffrey Chase
High-accuracy three-dimensional (3-D) breast-surface reconstruction is pivotal for vibration-based diagnostic modalities, such as the Digital Imaging Elasto-Tomography (DIET) breast cancer screening technology. Yet the coupled influence of geometric complexity and measurement noise in structured-light systems has not been quantified rigorously in this type of application. To address this gap, we establish a dual-track evaluation framework combining ground-truth numerical simulations with laser-scanned silicone phantom data to assess reconstruction fidelity under clinically relevant conditions.
Analytically defined hemispherical point clouds, both regular and irregular, were synthesised and corrupted with random and periodic noise of varying amplitudes, creating a benchmark data set with explicit, known ground truth. In parallel, anatomically realistic silicone breast phantoms without internal, high stiffness tumour-mimicking inclusions were imaged using a 39-beam, 520 nm structured-light system. Approximately 70 calibration poses were acquired per laser line, yielding sub-millimetre accuracy. All data were processed within a spherical-polar parameterisation and optimised by a Levenberg–Marquardt scheme. Reconstruction error was quantified via root-mean-square error (RMSE), inter-quartile range (IQR), outlier counts, and multiple visualisations (histograms, spatial error maps, and boxplots).
Results reveal a clear performance hierarchy. Under noise-free conditions the simulated hemisphere shows the lowest RMSE, whereas sparse sampling in the phantoms causes local surface depressions. Systematic (periodic) noise dominates the error budget, increasing RMSE by an order of magnitude and producing persistent ring- or band-shaped artefacts across all reconstructions. Random noise mainly introduces local high-frequency roughness. Concave deformations generate a stable toroidal residual whose contribution to global RMSE remains below 10 % in the absence of noise. When geometric irregularities and noise coexist, the resulting errors greatly exceed those due to shape alone, underscoring noise suppression as the primary determinant of best attainable accuracy in this application.
The proposed simulation-and-phantom framework delivers the first comprehensive error map for structured-light breast reconstruction in a DIET system and indicates future algorithms must integrate curvature-adaptive modelling with explicit compensation for periodic perturbations to remain robust across the wide spectrum of breast sizes and shapes encountered in clinical practice.
{"title":"Impact of geometric irregularity and error on 3D surface reconstruction in Digital Imaging Elasto-Tomography System","authors":"Yuwei Sun, Samantha Couper, Cong Zhou, J. Geoffrey Chase","doi":"10.1016/j.ifacsc.2025.100332","DOIUrl":"10.1016/j.ifacsc.2025.100332","url":null,"abstract":"<div><div>High-accuracy three-dimensional (3-D) breast-surface reconstruction is pivotal for vibration-based diagnostic modalities, such as the Digital Imaging Elasto-Tomography (DIET) breast cancer screening technology. Yet the coupled influence of geometric complexity and measurement noise in structured-light systems has not been quantified rigorously in this type of application. To address this gap, we establish a dual-track evaluation framework combining ground-truth numerical simulations with laser-scanned silicone phantom data to assess reconstruction fidelity under clinically relevant conditions.</div><div>Analytically defined hemispherical point clouds, both regular and irregular, were synthesised and corrupted with random and periodic noise of varying amplitudes, creating a benchmark data set with explicit, known ground truth. In parallel, anatomically realistic silicone breast phantoms without internal, high stiffness tumour-mimicking inclusions were imaged using a 39-beam, 520 nm structured-light system. Approximately 70 calibration poses were acquired per laser line, yielding sub-millimetre accuracy. All data were processed within a spherical-polar parameterisation and optimised by a Levenberg–Marquardt scheme. Reconstruction error was quantified via root-mean-square error (RMSE), inter-quartile range (IQR), outlier counts, and multiple visualisations (histograms, spatial error maps, and boxplots).</div><div>Results reveal a clear performance hierarchy. Under noise-free conditions the simulated hemisphere shows the lowest RMSE, whereas sparse sampling in the phantoms causes local surface depressions. Systematic (periodic) noise dominates the error budget, increasing RMSE by an order of magnitude and producing persistent ring- or band-shaped artefacts across all reconstructions. Random noise mainly introduces local high-frequency roughness. Concave deformations generate a stable toroidal residual whose contribution to global RMSE remains below 10 % in the absence of noise. When geometric irregularities and noise coexist, the resulting errors greatly exceed those due to shape alone, underscoring noise suppression as the primary determinant of best attainable accuracy in this application.</div><div>The proposed simulation-and-phantom framework delivers the first comprehensive error map for structured-light breast reconstruction in a DIET system and indicates future algorithms must integrate curvature-adaptive modelling with explicit compensation for periodic perturbations to remain robust across the wide spectrum of breast sizes and shapes encountered in clinical practice.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"33 ","pages":"Article 100332"},"PeriodicalIF":1.8,"publicationDate":"2025-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144907139","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-13DOI: 10.1016/j.ifacsc.2025.100331
Savin Treanţă , Cristina-Mihaela Cebuc
This paper investigates a distinctive family of (weak) composite controlled variational inequality models of vector type, contextualized within the domain of vector control. We establish its intrinsic relationships with the composite multiple objective variational control problem, shedding light on the intricate dynamics of control systems. An exploration of the connections between a critical or (weak) efficient point in the composite vector optimization problem and a solution of the associated composite vector controlled variational-like inequality is undertaken, guided by the assumption of composite (strictly) invariant convexity and/or pseudo-invariant convexity of the involved functionals. By using the KKM lemma, a result on the existence of solutions for the composite controlled variational inequality problem is stated. Additionally, a gap functional is defined specifically for the composite controlled variational inequality problem, providing a valuable tool for understanding the nuances of control scenarios. The significance of our findings is exemplified through illustrative examples, offering concrete insights into applying the derived results in vector control scenarios.
{"title":"On some composite multiple-cost models and associated vector inequalities","authors":"Savin Treanţă , Cristina-Mihaela Cebuc","doi":"10.1016/j.ifacsc.2025.100331","DOIUrl":"10.1016/j.ifacsc.2025.100331","url":null,"abstract":"<div><div>This paper investigates a distinctive family of (weak) composite controlled variational inequality models of vector type, contextualized within the domain of vector control. We establish its intrinsic relationships with the composite multiple objective variational control problem, shedding light on the intricate dynamics of control systems. An exploration of the connections between a critical or (weak) efficient point in the composite vector optimization problem and a solution of the associated composite vector controlled variational-like inequality is undertaken, guided by the assumption of composite (strictly) invariant convexity and/or pseudo-invariant convexity of the involved functionals. By using the KKM lemma, a result on the existence of solutions for the composite controlled variational inequality problem is stated. Additionally, a gap functional is defined specifically for the composite controlled variational inequality problem, providing a valuable tool for understanding the nuances of control scenarios. The significance of our findings is exemplified through illustrative examples, offering concrete insights into applying the derived results in vector control scenarios.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"33 ","pages":"Article 100331"},"PeriodicalIF":1.8,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144864588","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-08-07DOI: 10.1016/j.ifacsc.2025.100330
Eleni Zavrakli , Andrew Parnell , Subhrakanti Dey
With the rapid development of Additive Manufacturing (AM) comes an urgent need for advanced monitoring and control of the process. Many aspects of the AM process play a significant role in the efficiency, accuracy, and repeatability of the process, with temperature regulation being one of the most important. In this work, we solve the problem of optimal tracking control for a state space temperature model of a Big Area Additive Manufacturing (BAAM) system. In particular, we address the problem of designing a Linear Quadratic Tracking (LQT) controller when access to the exact system state is not possible, except in the form of measurements. We initially solve the problem with a model-based approach based on reinforcement learning concepts with state estimation through an observer. We then design a model-free reinforcement-learning based controller with an internal state estimation step and demonstrate its performance through a simulator of the systems’ behaviour. In addition, we explore the use of Bayesian Optimisation as a means to optimise the design parameters of the LQT problem. Our results showcase the possibility of achieving optimal behaviour while learning directly from process data, independently of a model of the process. This is an encouraging outcome towards the realisation of intelligent manufacturing.
{"title":"Reinforcement Learning based temperature regulation for a Material Extrusion Additive Manufacturing system","authors":"Eleni Zavrakli , Andrew Parnell , Subhrakanti Dey","doi":"10.1016/j.ifacsc.2025.100330","DOIUrl":"10.1016/j.ifacsc.2025.100330","url":null,"abstract":"<div><div>With the rapid development of Additive Manufacturing (AM) comes an urgent need for advanced monitoring and control of the process. Many aspects of the AM process play a significant role in the efficiency, accuracy, and repeatability of the process, with temperature regulation being one of the most important. In this work, we solve the problem of optimal tracking control for a state space temperature model of a Big Area Additive Manufacturing (BAAM) system. In particular, we address the problem of designing a Linear Quadratic Tracking (LQT) controller when access to the exact system state is not possible, except in the form of measurements. We initially solve the problem with a model-based approach based on reinforcement learning concepts with state estimation through an observer. We then design a model-free reinforcement-learning based controller with an internal state estimation step and demonstrate its performance through a simulator of the systems’ behaviour. In addition, we explore the use of Bayesian Optimisation as a means to optimise the design parameters of the LQT problem. Our results showcase the possibility of achieving optimal behaviour while learning directly from process data, independently of a model of the process. This is an encouraging outcome towards the realisation of intelligent manufacturing.</div></div>","PeriodicalId":29926,"journal":{"name":"IFAC Journal of Systems and Control","volume":"33 ","pages":"Article 100330"},"PeriodicalIF":1.8,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144810510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号