Search

Complex robotic systems, such as humanoid robot hands, soft robots, and walking robots, pose a challenging control problem due to their high dimensionality and heavy non-linearities. Conventional model-based feedback controllers demonstrate robustness and stability but struggle to cope with the escalating system design and tuning complexity accompanying larger dimensions. In contrast, data-driven methods such as artificial neural networks excel at representing high-dimensional data but lack robustness, generalization, and real-time adaptiveness. In response to these challenges, researchers are directing their focus to biological paradigms, drawing inspiration from the remarkable control capabilities inherent in the human body. This has motivated the exploration of new control methods aimed at closely emulating the motor functions of the brain given the current insights in neuroscience. Recent investigation into these Brain-Inspired control techniques have yielded promising results, notably in tasks involving trajectory tracking and robot locomotion. This paper presents a comprehensive review of the foremost trends in biomimetic brain-inspired controlmethods to tackle the intricacies associated with controlling complex robotic systems. [ABSTRACT FROM AUTHOR]

Gait recognition is an advanced biometric modality that identifies individuals based on their walking postures. Currently, the majority of gait recognition methods employ silhouette images as the primary representation of gait. However, the inclusion of silhouette information, which incorporates details about clothing and carried items, may interfere with the extraction of gait features. To this end, we present GaitSkeleton, a novel approach that utilizes skeleton sequences as the gait representation. GaitSkeleton models the motor coordination feature of gait and combines three-scale spatial features to obtain discriminative features for gait recognition. To capture the coordination relationship between joints, we design a Coordination Feature Learner (CFL). CFL is embedded into the ST-GCN block to enhance the model's representational capacity. Additionally, we introduce a novel feature cascade module called MSCM, which further enhances the discriminative information by sampling spatial features at three scales. The entire network significantly enhances the discriminative nature of gait features. Experimental results on the CASIA-B dataset demonstrate that our method achieves state-of-the-art performance in model-based gait recognition. [ABSTRACT FROM AUTHOR]

Dopamine release in the nucleus accumbens core (NAcC) is generally considered to be a proxy for phasic firing of the ventral tegmental area dopamine (VTADA) neurons. Thus, dopamine release in NAcC is hypothesized to reflect a unitary role in reward prediction error signaling. However, recent studies reveal more diverse roles of dopamine neurons, which support an emerging idea that dopamine regulates learning differently in distinct circuits. To understand whether the NAcC might regulate a unique component of learning, we recorded dopamine release in NAcC while male rats performed a backward conditioning task where a reward is followed by a neutral cue. We used this task because we can delineate different components of learning, which include sensory-specific inhibitory and general excitatory components. Furthermore, we have shown that VTADA neurons are necessary for both the specific and general components of backward associations. Here, we found that dopamine release in NAcC increased to the reward across learning while reducing to the cue that followed as it became more expected. This mirrors the dopamine prediction error signal seen during forward conditioning and cannot be accounted for temporal-difference reinforcement learning. Subsequent tests allowed us to dissociate these learning components and revealed that dopamine release in NAcC reflects the general excitatory component of backward associations, but not their sensory-specific component. These results emphasize the importance of examining distinct functions of different dopamine projections in reinforcement learning. [ABSTRACT FROM AUTHOR]

Digital engineering and digital twins are increasingly being used in nuclear energy projects with important impacts. At Idaho National Laboratory, these approaches have been applied in a variety of nuclear energy research, development, and demonstration projects, with key lessons and evolutions occurring for each. In this paper, we describe the use of digital engineering and digital twins in the Versatile Test Reactor design, National Reactor Innovation Center test beds, and nonproliferation analysis of the AGN-201 reactor design. We share key lessons learned for these projects related to tool selection, adoption and training, and working with existing assets versus beginning at the design phase. We also share highlights of future potential uses of digital twins and digital engineering, including using artificial intelligence to perform repetitive design tasks and digital twins to move towards semiautonomous nuclear power plant operations. [ABSTRACT FROM AUTHOR]

While previous research has shown that compulsivity is related to an imbalance between goal‐directed and habitual learning systems, very little is known about whether this effect is due to the impairment of a single system or the impairment of the arbitration mechanism that determines which system controls behaviour at any given moment; the current study aims to address this disagreement. Nineteen alcohol use disorder, 30 obsessive‐compulsive disorder (OCD) and 20 major depressive disorder patients and corresponding sex‐ and age‐matched controls performed two‐choice, three‐stage Markov decision‐making paradigm. Model‐based and mode‐free reinforcement learning models were used to independently fitted their behavioural data. Alcohol use disorder and OCD patients showed less model‐based strategy choice than healthy controls in task conditions where the model‐based strategy was optimal. Only OCD patients showed higher behavioural control system switching in task conditions where model‐free use was optimal. Major depressive disorder patients did not differ from the matched control in both. These findings suggest that dysfunction in arbitration control between dual systems may be the basis for diverse disorders involving compulsivity. [ABSTRACT FROM AUTHOR]

Abstract Wear of the ultra-high molecular-weight polyethylene (UHMWPE) component in total knee arthroplasty contributes to implant failure. It is often detected late, when patients experience pain or instability. Early monitoring could enable timely intervention, preventing implant failure and joint degeneration. This study investigates the accuracy and precision (repeatability) of model-based wear measurement (MBWM), a novel technique that can estimate inlay thickness and wear radiographically. Six inlays were milled from non-crosslinked UHMWPE and imaged via X-ray in anteroposterior view at flexion angles 0°, 30°, and 60° on a phantom knee model. MBWM measurements were compared with reference values from a coordinate measurement machine. Three inlays were subjected to accelerated wear generation and similarly evaluated. MBWM estimated inlay thickness with medial and lateral accuracies of 0.13 ± 0.09 and 0.14 ± 0.09 mm, respectively, and linear wear with an accuracy of 0.07 ± 0.06 mm. Thickness measurements revealed significant lateral differences at 0° and 30° (0.22 ± 0.08 mm vs. 0.06 ± 0.06 mm, respectively; t-test, p = 0.0002). Precision was high, with average medial and lateral differences of − 0.01 ± 0.04 mm between double experiments. MBWM using plain radiographs presents a practical and promising approach for the clinical detection of implant wear.

Introduction The global measles incidence has decreased from 145 to 49 cases per 1 million population from 2000 to 2018, but evaluating the economic benefits of a second measles-containing vaccine (MCV2) is crucial. This study reviewed the evidence and quality of economic evaluation studies to guide MCV2 introduction.Methods The systematic review of model-based economic evaluation studies was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. The search yielded 2231 articles, with 876 duplicates removed and 1355 articles screened, with nine studies included for final analysis.Results Six studies reported a positive benefit–cost ratio with one resulting in net savings of $11.6 billion, and two studies estimated a 2-dose MMR vaccination program would save $119.24 to prevent one measles case, and a second dose could prevent 9,200 cases at 18 months, saving $548.19 per case. The most sensitive variables were the discount rate and vaccination administration cost.Conclusions Two MCV doses or a second opportunity with an additional dose of MCV were highly cost-beneficial and resulted in substantial cost savings compared to a single routine vaccine. But further research using high-quality model-based health economic evaluation studies of MCV2 should be made available to decision-makers.PROSPERO Registration CRD42020200669

There are very few integration methods satisfying the explicit displacement formulation, unconditional stability, and higher-order accuracy, simultaneously. This paper proposes a new predictor–corrector explicit (NPCE) integration method with unconditional stability and higher-order accuracy. The NPCE is developed by matching its temporal discrete equation (TDE) with that of a recently developed fourth-order new dual-explicit (NDE) integration method. The accuracy and stability of the NPCE scheme are analyzed. It is found that the NPCE method possesses the same stability property as the NDE method. In addition, the NPCE method can achieve the same accuracy level as the NDE method for the undamped system, and it is more accurate than the NDE method for the damped system. The performance of the NPCE method is illustrated by four numerical examples. Some representative integration methods are adopted for comparison. The results indicate that the NPCE method has excellent accuracy with low relative absolute and squared errors. [ABSTRACT FROM AUTHOR]

The development of safety-critical real-time embedded systems (RTESs) is a difficult task. Indeed, in addition to functional requirements, these systems must guarantee timing constraints. Model-driven engineering (MDE) promotes a rise in the level of abstraction during the development process of these systems. This is beneficial since it may help designers manage the increasing complexity of these systems. It may also accelerate the development process through model transformations and enable early verification of the timing properties. Schedulability analysis techniques aim to verify whether the tasks complete their executions within the time limit specified by the real-time application (i.e., the deadline). Unfortunately, software designers find this task difficult since it requires a deep understanding of real-time scheduling theory, as well as the use of dedicated tools (called analysis tools). In this paper, we propose real-time verification design patterns to promote early verification of timing properties in a model-based approach. These patterns aim to assist designers in the verification process. They encapsulate the knowledge of a real-time expert in scheduling theory. To facilitate and fasten the verification process, we propose to automate the generation of the analysis model. Thanks to the provided patterns, this generation phase includes an automatic consistency check of the design model. The consistency check step verifies the design's validity and completeness with respect to the applied pattern. To add further benefits to the proposed patterns, we describe a process that provides context for how these patterns can be applied. We apply and simulate the contribution to real-world case study. The obtained results show the viability of this research. [ABSTRACT FROM AUTHOR]

Complex robotic systems, such as humanoid robot hands, soft robots, and walking robots, pose a challenging control problem due to their high dimensionality and heavy non-linearities. Conventional model-based feedback controllers demonstrate robustness and stability but struggle to cope with the escalating system design and tuning complexity accompanying larger dimensions. In contrast, data-driven methods such as artificial neural networks excel at representing high-dimensional data but lack robustness, generalization, and real-time adaptiveness. In response to these challenges, researchers are directing their focus to biological paradigms, drawing inspiration from the remarkable control capabilities inherent in the human body. This has motivated the exploration of new control methods aimed at closely emulating the motor functions of the brain given the current insights in neuroscience. Recent investigation into these Brain-Inspired control techniques have yielded promising results, notably in tasks involving trajectory tracking and robot locomotion. This paper presents a comprehensive review of the foremost trends in biomimetic brain-inspired control methods to tackle the intricacies associated with controlling complex robotic systems.

Digital engineering and digital twins are increasingly being used in nuclear energy projects with important impacts. At Idaho National Laboratory, these approaches have been applied in a variety of nuclear energy research, development, and demonstration projects, with key lessons and evolutions occurring for each. In this paper, we describe the use of digital engineering and digital twins in the Versatile Test Reactor design, National Reactor Innovation Center test beds, and nonproliferation analysis of the AGN-201 reactor design. We share key lessons learned for these projects related to tool selection, adoption and training, and working with existing assets versus beginning at the design phase. We also share highlights of future potential uses of digital twins and digital engineering, including using artificial intelligence to perform repetitive design tasks and digital twins to move towards semiautonomous nuclear power plant operations.

In the automotive industry, the accurate estimation of wheel displacements is crucial for optimizing vehicle suspension systems. Traditional model-based approaches often face challenges in accurately predicting these displacements due to the complex dynamics of the road-vehicle interaction. To address this limitation, this study, conducted in the frame of the OWHEEL project, proposes the integration of a multi-output neural network capable of compensating for estimation errors inherent in model-based approaches, specifically those arising from road inputs. Leveraging only vertical acceleration measurements, the neural network operates in parallel with the model-based estimator, enhancing the overall accuracy of displacement estimation. Experimental validation using a sports vehicle demonstrates the efficacy of the proposed methodology, showcasing its ability to improve estimation accuracy beyond the capabilities of the model-based approach alone.

This study proposes a hybrid fault detection methodology for detecting epoxy faults in hairpin-based stator windings of electric motors. The hybrid methodology integrates a model-based approach for feature extraction and a data-driven approach for fault classification. A new lumped-parameter equivalent circuit model specifically for hairpin windings is developed. It can accurately simulate the high-frequency impedance behaviors of hairpin windings and physically interpret the distinction of the measurement curves under different epoxy configurations. Using system identification, the parameters of this new model are identified to extract the features of phase windings, reflecting different fault conditions by varying the parameters in distinct ranges. Fault classification is implemented using a data-driven method to distinguish the underlying patterns, which is difficult to achieve by conventional threshold limit checking due to the inevitably introduced noise and uncertainties. Principal Component Analysis (PCA) is applied to refine the features, followed by a Support Vector Machine (SVM) performing fault classification. The proposed hybrid methodology successfully detects epoxy-related fault conditions, providing a new strategy for fault detection.

When a facial image is blurred, it significantly affects high-level vision tasks such as face recognition. The purpose of facial image deblurring is to recover a clear image from a blurry input image, which can improve the recognition accuracy, etc. However, general deblurring methods do not perform well on facial images. Therefore, some face deblurring methods have been proposed to improve performance by adding semantic or structural information as specific priors according to the characteristics of the facial images. In this paper, we survey and summarize recently published methods for facial image deblurring, most of which are based on deep learning. First, we provide a brief introduction to the modeling of image blurring. Next, we summarize face deblurring methods into two categories: model-based methods and deep learning-based methods. Furthermore, we summarize the datasets, loss functions, and performance evaluation metrics commonly used in the neural network training process. We show the performance of classical methods on these datasets and metrics and provide a brief discussion on the differences between model-based and learning-based methods. Finally, we discuss the current challenges and possible future research directions. [ABSTRACT FROM AUTHOR]

People with methamphetamine use disorder (MUD) struggle to shift their behaviour from methamphetamine‐orientated habits to goal‐oriented choices. The model‐based/model‐free framework is well suited to understand this difficulty by unpacking the computational mechanisms that support experienced‐based (model‐free) and goal‐directed (model‐based) choices. We aimed to examine whether 1) participants with MUD differed from controls on behavioural proxies and/or computational mechanisms of model‐based/model‐free choices; 2) model‐based/model‐free decision‐making correlated with MUD symptoms; and 3) model‐based/model‐free deficits improved over six weeks in the group with MUD. Participants with MUD and controls with similar age, IQ and socioeconomic status completed the Two‐Step Task at treatment commencement (MUD n = 30, Controls n = 31) and six weeks later (MUD n = 23, Controls n = 26). We examined behavioural proxies of model‐based/model‐free decisions using mixed logistic regression, and their underlying mechanisms using computational modelling. At a behavioural level, participants with MUD were more likely to switch their choices following rewarded actions, although this pattern improved at follow up. At a computational level, groups were similar in their use of model‐based mechanisms, but participants with MUD were less likely to apply model‐free mechanisms and less likely to repeat rewarded actions. We did not find evidence that individual differences in model‐based or model‐free parameters were associated with greater severity of methamphetamine dependence, nor did we find that group differences in computational parameters changed between baseline and follow‐up assessment. Decision‐making challenges in people with MUD are likely related to difficulties in pursuing choices previously associated with positive outcomes. [ABSTRACT FROM AUTHOR]

Applying remote sensing models outside the temporal range of their training data, referred to as temporal model transfer, has become common practice for large area monitoring projects that extrapolate models for hindcasting or forecasting to time periods lacking reference data. However, the development of appropriate validation methods for temporal transfer has lagged behind its rapid adoption. Breaking temporal transfer's assumption of temporal consistency in both remote sensing and reference data and their relationship to each other could lead to biased pixel-level predictions and small area estimators, compromising the operational validity of large area monitoring products. Few studies using temporal transfer have evaluated its effects on model accuracy at the pixel/plot level, and the propensity for biased small area estimators and trends resulting from temporal transfer remains unexplored. We present a framework for evaluating temporal transferability by combining temporal cross-validation with a multiscale map assessment to aid in identifying where and when biased model predictions could scale to small area estimates and affect predicted trends.This validation framework is demonstrated in a case study of annual percent tree canopy cover mapping in Michigan. We tested and compared temporal transferability of random forest models of canopy cover derived from 2010 to 2016 systematic dot-grid photo-interpretations at Forest Inventory and Analysis plots with Landsat spectral indices fit with the LandTrendr temporal segmentation algorithm serving as the primary predictor variables. The temporal cross-validation error (mean RMSE = 13.9% cover) was higher than the common validation approach of considering all time periods of testing data together (RMSE = 13.6% cover) and lower than models trained and tested within the same year (mean RMSE = 14.2% cover). However, the bias of model predictions and small area estimators for individual years was higher with temporal transfer models than when applying models within the same year as their training data. We also evaluated how training models using different temporal subsets and with and without LandTrendr fitting affected predictions of Michigan's 1984–2020 predicted annual mean cover. The mean cover from LandTrendr-based models followed expected and consistent trends and had less difference between models trained with different temporal subsets (max difference = 5.8% cover). While those from Landsat had high interannual variations and greater difference between temporal models (max difference = 11.2% cover). The results of this case study demonstrate that evaluation of temporal transferability is necessary for establishing the operational validity of large area monitoring products, even when using time series algorithms that improve temporal consistency.

Abstract When a facial image is blurred, it significantly affects high-level vision tasks such as face recognition. The purpose of facial image deblurring is to recover a clear image from a blurry input image, which can improve the recognition accuracy, etc. However, general deblurring methods do not perform well on facial images. Therefore, some face deblurring methods have been proposed to improve performance by adding semantic or structural information as specific priors according to the characteristics of the facial images. In this paper, we survey and summarize recently published methods for facial image deblurring, most of which are based on deep learning. First, we provide a brief introduction to the modeling of image blurring. Next, we summarize face deblurring methods into two categories: model-based methods and deep learning-based methods. Furthermore, we summarize the datasets, loss functions, and performance evaluation metrics commonly used in the neural network training process. We show the performance of classical methods on these datasets and metrics and provide a brief discussion on the differences between model-based and learning-based methods. Finally, we discuss the current challenges and possible future research directions.

Abstract This article presents the development of the Control‐oriented, Reconfigurable, and Acausal Floating Turbine Simulator (CRAFTS). CRAFTS has a modular, hierarchical model architecture that enables rapid and accurate simulation of wind turbines. The architecture facilitates the incorporation of model variants, and its system reconfiguration features help simulate multiple design variants. CRAFTS also supports the integration of models developed on a causality‐free platform (e.g., Modelica®) with existing causal models. This article focuses on the validation of a land‐based wind turbine models against OpenFAST, an industry‐standard platform, for several test cases. Closed‐loop scenarios are also tested using the standard ROSCO controller and compared against OpenFAST. In addition, a nonlinear controller developed in our prior work is also evaluated. The test cases demonstrate the user‐friendly and computationally efficient capabilities of CRAFTS to facilitate control co‐design, assist in incorporating multiphysics models, be adaptable to design variants, and allow for rapid simulations to validate models and evaluate controllers.

The application of FPGA is becoming wider because of its advantages of reconfigurability, high performance, low power consumption, and strong real-time performance. However, its shortcomings such as high development threshold, low efficiency, and long development cycle become more prominent as the increase of system size and complexity. This paper uses model-based development methods to study the correlation algorithm of FPGA, including Simulink model building, test validation, code generation and simulation. The results of comparison with traditional manual coding show that the FPGA software development cycle using model-based design methods is only one fifth of the original method and the error in calculation results of the correlation coefficient is also under 0.3%. Although the length of generated code far exceeds that of handwritten code, the resource utilization of logic units, register, storage units, and multiplier units, etc. at runtime are less than those of handwritten code after code synthesis. It can be seen that model-based development methods not only meet the accuracy requirements (not exceeding ±5‰), but also greatly shorten the development cycle and lower the coding threshold.

This article presents the development of the Control‐oriented, Reconfigurable, and Acausal Floating Turbine Simulator (CRAFTS). CRAFTS has a modular, hierarchical model architecture that enables rapid and accurate simulation of wind turbines. The architecture facilitates the incorporation of model variants, and its system reconfiguration features help simulate multiple design variants. CRAFTS also supports the integration of models developed on a causality‐free platform (e.g., Modelica®) with existing causal models. This article focuses on the validation of a land‐based wind turbine models against OpenFAST, an industry‐standard platform, for several test cases. Closed‐loop scenarios are also tested using the standard ROSCO controller and compared against OpenFAST. In addition, a nonlinear controller developed in our prior work is also evaluated. The test cases demonstrate the user‐friendly and computationally efficient capabilities of CRAFTS to facilitate control co‐design, assist in incorporating multiphysics models, be adaptable to design variants, and allow for rapid simulations to validate models and evaluate controllers. [ABSTRACT FROM AUTHOR]

Decarbonization of the transport sector could be achieved through fuel cell technology. The candidature of this technology is motivated by its high current density and lack of emissions. However, its widespread deployment is restrained by durability and reliability constraints. During normal operation, the fuel cell system supplies stable power to the load. Contrarily, when it is operated under faulty conditions, the system's output power deteriorates, leading to low durability. It is therefore of paramount importance to ensure that the system is operated in a non-faulty condition. In this paper, we provide a critical review of the analytical-redundancy-based fault diagnosis methods for proton exchange membrane fuel cells (PEMFCs). An in-depth analysis of the various methods has been presented in terms of accuracy, complexity, implementability, and robustness to aging and dynamic operating conditions. [ABSTRACT FROM AUTHOR]

The bacterium Escherichia coli is a widely used organism in biotechnology. For high space-time yields, glucose-limited fed-batch technology is the industry standard; this is because an overflow metabolism of acetate occurs at high glucose concentrations. As an interesting alternative, various strains with limited glucose uptake have been developed. However, these have not yet been characterized under process conditions. To demonstrate the efficiency of our previously developed high-throughput robotic platform, in the present work, we characterized three different exemplary E. coli knockout (KO) strains with limited glucose uptake capacities at three different scales (microtiter plates, 10 mL bioreactor system and 100 mL bioreactor system) under excess glucose conditions with different initial glucose concentrations. The extensive measurements of growth behavior, substrate consumption, respiration, and overflow metabolism were then used to determine the appropriate growth parameters using a mechanistic mathematical model, which allowed for a comprehensive comparative analysis of the strains. The analysis was performed coherently with these different reactor configurations and the results could be successfully transferred from one platform to another. Single and double KO mutants showed reduced specific rates for substrate uptake qSmax and acetate production qApmax; meanwhile, higher glucose concentrations had adverse effects on the biomass yield coefficient YXSem. Additional parameters compared to previous studies for the oxygen uptake rate and carbon dioxide production rate indicated differences in the specific oxygen uptake rate qOmax. This study is an example of how automated robotic equipment, together with mathematical model-based approaches, can be successfully used to characterize strains and obtain comprehensive information more quickly, with a trade-off between throughput and analytical capacity. [ABSTRACT FROM AUTHOR]

Fundamental to most classical data collection sampling theory development is the random drawings assumption requiring that each targeted population member has a known sample selection (i.e., inclusion) probability. Frequently, however, unrestricted random sampling of spatially autocorrelated data is impractical and/or inefficient. Instead, randomly choosing a population subset accounts for its exhibited spatial pattern by utilizing a grid, which often provides improved parameter estimates, such as the geographic landscape mean, at least via its precision. Unfortunately, spatial autocorrelation latent in these data can produce a questionable mean and/or standard error estimate because each sampled population member contains information about its nearby members, a data feature explicitly acknowledged in model-based inference, but ignored in design-based inference. This autocorrelation effect prompted the development of formulae for calculating an effective sample size (i.e., the equivalent number of sample selections from a geographically randomly distributed population that would yield the same sampling error) estimate. Some researchers recently challenged this and other aspects of spatial statistics as being incorrect/invalid/misleading. This paper seeks to address this category of misconceptions, demonstrating that the effective geographic sample size is a valid and useful concept regardless of the inferential basis invoked. Its spatial statistical methodology builds upon the preceding ingredients.

Aviation is characterized by many stakeholders, long lifespans of its assets, and high requirements regarding safety, security, and documentation. To meet these requirements as well as customer needs, aircraft are regularly retrofitted with new cabins. During the planning and execution of this cabin retrofit, handling the needed and available data poses a challenge to the engineers. While much of the required data is available in some form, generally there is a lack of a digitally usable dataset of the specific aircraft—a virtual representation of the physical asset is missing. To support the implementation of such a digital twin and, thus, the overall process of retrofitting aircraft, an approach to model-driven data handling tailored to the unique circumstances and requirements of aviation is introduced. The methodology consists of a combination of systems engineering and data science techniques framed by an overarching procedure that iteratively creates and enhances a digitally accessible dataset of the relevant data, hence supporting the retrofit engineers by easing access to needed information. Besides the presentation of the research background and the methodology, a simplified example is shown, demonstrating the approach using abstracted but realistic information provided by partners from the industry. [ABSTRACT FROM AUTHOR]

The deployment of electro-mechanical actuators plays an important role towards the adoption of the more electric aircraft (MEA) philosophy. On the other hand, a seamless substitution of EMAs, in place of more traditional hydraulic solutions, is still set back, due to the shortage of real-life and reliability data regarding their failure modes. One way to work around this problem is providing a capillary EMA prognostics and health management (PHM) system capable of recognizing failures before they actually undermine the ability of the safety-critical system to perform its functions. The aim of this work is the development of a model-based prognostic framework for PMSM-based EMAs leveraging a metaheuristic algorithm: the evolutionary (differential evolution (DE)) and swarm intelligence (particle swarm (PSO), grey wolf (GWO)) methods are considered. Several failures (dry friction, backlash, short circuit, eccentricity, and proportional gain) are simulated by a reference model, and then detected and identified by the envisioned prognostic method, which employs a low fidelity monitoring model. The paper findings are analysed, showing good results and proving that this strategy could be executed and integrated in more complex routines, supporting EMAs adoption, with positive impacts on system safety and reliability in the aerospace and industrial field. [ABSTRACT FROM AUTHOR]

When a facial image is blurred, it significantly affects high-level vision tasks such as face recognition. The purpose of facial image deblurring is to recover a clear image from a blurry input image, which can improve the recognition accuracy, etc. However, general deblurring methods do not perform well on facial images. Therefore, some face deblurring methods have been proposed to improve performance by adding semantic or structural information as specific priors according to the characteristics of the facial images. In this paper, we survey and summarize recently published methods for facial image deblurring, most of which are based on deep learning. First, we provide a brief introduction to the modeling of image blurring. Next, we summarize face deblurring methods into two categories: model-based methods and deep learning-based methods. Furthermore, we summarize the datasets, loss functions, and performance evaluation metrics commonly used in the neural network training process. We show the performance of classical methods on these datasets and metrics and provide a brief discussion on the differences between model-based and learning-based methods. Finally, we discuss the current challenges and possible future research directions. [ABSTRACT FROM AUTHOR]

In this paper, leak detection and localization in water distribution networks will be reviewed. In particular, the paper presents the evolution of the methods from model-based towards data-based approaches, listing, describing and comparing the main and/or most recent methods of both categories. Besides, the practical applicability in real water utilities of different existing methods is discussed, outlining the advantages and limitations of model-based and data-driven methods for this task. A well-known case study is used to compare some of the more promising methods and illustrate their performances. Perspectives of the future evolution of the current existing methods are also provided. • A review of the most important leak management methods in water networks is presented. • Both model-based and data-driven approaches are applied to the BattLeDIM2020 benchmark. • General conclusions about the current state of the leak management field are extracted. • Guidelines regarding the interest of water utilities are provided. [ABSTRACT FROM AUTHOR]

Dopamine (DA) neurons in the ventral tegmental area (VTA) and substantia nigra (SNc) encode reward prediction errors (RPEs) and are proposed to mediate error-driven learning. However, the learning strategy engaged by DA-RPEs remains controversial. RPEs might imbue predictive cues with pure value, independently of representations of their associated outcome. Alternatively, RPEs might promote learning about the sensory features (the identity) of the rewarding outcome. Here, we show that, although both VTA and SNc DA neuron activation reinforces instrumental responding, only VTA DA neuron activation during consumption of expected sucrose reward restores error-driven learning and promotes formation of a new cue→sucrose association. Critically, expression of VTA DA-dependent Pavlovian associations is abolished following sucrose devaluation, a signature of identity-based learning. These findings reveal that activation of VTA- or SNc-DA neurons engages largely dissociable learning processes with VTA-DA neurons capable of participating in outcome-specific predictive learning, and the role of SNc-DA neurons appears limited to reinforcement of instrumental responses.

This review paper looks briefly at conventional approaches and examines the intelligent means for fault diagnosis (FD) and condition monitoring (CM) of electrical drives in detail, especially the ones that are common in Industry 4.0. After giving an overview on fault statistics, standard methods for the FD and CM of rotating machines are first visited, and then its orientation towards intelligent approaches is discussed. Major diagnostic procedures are addressed in detail together with their advancements to date. In particular, the emphasis is given to motor current signature analysis (MCSA) and digital signal processing techniques (DSPTs) mostly used for feature engineering. Consequently, the statistical procedures and machine learning techniques (stemming from artificial intelligence—AI) are also visited to describe how FD is carried out in various systems. The effectiveness of the amalgamation of the model, signal, and data-based techniques for the FD and CM of inductions motors (IMs) is also highlighted in this review. It is worth mentioning that a variety of neural- and non-neural-based approaches are discussed concerning major faults in rotating machines. Finally, after a thorough survey of the diagnostic techniques based on specific faults for electrical drives, several open problems are identified and discussed. The paper concludes with important recommendations on where to divert the research focus considering the current advancements in the FD and CM of rotating machines. [ABSTRACT FROM AUTHOR]

Fundamental to most classical data collection sampling theory development is the random drawings assumption requiring that each targeted population member has a known sample selection (i.e., inclusion) probability. Frequently, however, unrestricted random sampling of spatially autocorrelated data is impractical and/or inefficient. Instead, randomly choosing a population subset accounts for its exhibited spatial pattern by utilizing a grid, which often provides improved parameter estimates, such as the geographic landscape mean, at least via its precision. Unfortunately, spatial autocorrelation latent in these data can produce a questionable mean and/or standard error estimate because each sampled population member contains information about its nearby members, a data feature explicitly acknowledged in model-based inference, but ignored in design-based inference. This autocorrelation effect prompted the development of formulae for calculating an effective sample size (i.e., the equivalent number of sample selections from a geographically randomly distributed population that would yield the same sampling error) estimate. Some researchers recently challenged this and other aspects of spatial statistics as being incorrect/invalid/misleading. This paper seeks to address this category of misconceptions, demonstrating that the effective geographic sample size is a valid and useful concept regardless of the inferential basis invoked. Its spatial statistical methodology builds upon the preceding ingredients. [ABSTRACT FROM AUTHOR]

• The new algorithm is explicit for both displacement and velocity. • The new algorithm is unconditionally stable for linear and nonlinear softening stiffness systems. • The new algorithm has extremely small period error. • The new algorithm has no energy dissipation. • There still exist overshoot of the new algorithm. Several model-based or structure-dependent integration algorithms with explicit displacement formulation and unconditional stability are recently developed. According to the difference equations, these algorithms can be divided into dual-explicit and semi-explicit, where the first category algorithms are explicit for both displacement and velocity, while the second category are only explicit for displacement. The well-known Chen-Ricles (CR) and Chang algorithms are two representatives of dual-explicit and semi-explicit model-based algorithms, respectively. The model-based integration algorithms are promising with excellent computing efficiency, but most of them are second-order accurate. In this paper, a new dual-explicit (NDE) algorithm with higher-order accuracy is developed. The integration parameters of the NDE algorithm are obtained by using the pole mapping and second-order Pade approximation. Numerous important numerical characteristics including the consistency and local truncation error, stability, numerical dispersion and energy dissipation and overshoot are systematically analyzed for the new algorithm. It is found that the NDE algorithm is fourth-order accurate for free vibration without damping. As for forced vibration with/without damping and free vibration with damping, the NDE algorithm is no longer fourth-order accurate, but it is still more accurate than the CR and Chang algorithms. In addition, the period error of the algorithm is extremely small. Three representative numerical examples are finally provided to demonstrate the high accuracy of the NDE algorithm. Several classical and latest model-based integration algorithms and three fourth-order accurate algorithms are also adopted for comparisons. It has been proved that the NDE algorithm has much better performance than other selected model-based algorithms. [ABSTRACT FROM AUTHOR]

This literature review extends and contributes to research on the development of data-driven optimal control. Previous reviews have documented the development of model-based and data-driven control in isolation and have not critically reviewed reinforcement learning approaches for adaptive data-driven optimal control frameworks. The presented review discusses the development of model-based to model-free adaptive controllers, highlighting the use of data in control frameworks. In data-driven control frameworks, reinforcement learning methods may be used to derive the optimal policy for dynamical systems. Attractive characteristics of these methods include not requiring a mathematical model of complex systems, their inherent adaptive control capabilities, being an unsupervised learning technique and their decision-making abilities, which are both an advantage and motivation behind this approach. This review considers previous reviews on these topics, including recent work on data-driven control methods. In addition, this review shows the use of data to derive system dynamics, determine the control policy using feedback information, and tune fixed controllers. Furthermore, the review summarises various data-driven methods and their corresponding characteristics. Finally, the review provides a taxonomy, a timeline and a concise narrative of the development of model-based to model-free data-driven adaptive control and underlines the limitations of these techniques due to the lack of theoretical analysis. Areas of further work include theoretical analysis on stability and robustness for data-driven control systems, explainability of black-box policy learning techniques and an evaluation of the impact of the extension of system simulators to include digital twins.

The bacterium Escherichia coli is a widely used organism in biotechnology. For high space-time yields, glucose-limited fed-batch technology is the industry standard; this is because an overflow metabolism of acetate occurs at high glucose concentrations. As an interesting alternative, various strains with limited glucose uptake have been developed. However, these have not yet been characterized under process conditions. To demonstrate the efficiency of our previously developed high-throughput robotic platform, in the present work, we characterized three different exemplary E. coli knockout (KO) strains with limited glucose uptake capacities at three different scales (microtiter plates, 10 mL bioreactor system and 100 mL bioreactor system) under excess glucose conditions with different initial glucose concentrations. The extensive measurements of growth behavior, substrate consumption, respiration, and overflow metabolism were then used to determine the appropriate growth parameters using a mechanistic mathematical model, which allowed for a comprehensive comparative analysis of the strains. The analysis was performed coherently with these different reactor configurations and the results could be successfully transferred from one platform to another. Single and double KO mutants showed reduced specific rates for substrate uptake qSmax and acetate production qApmax; meanwhile, higher glucose concentrations had adverse effects on the biomass yield coefficient YXSem. Additional parameters compared to previous studies for the oxygen uptake rate and carbon dioxide production rate indicated differences in the specific oxygen uptake rate qOmax. This study is an example of how automated robotic equipment, together with mathematical model-based approaches, can be successfully used to characterize strains and obtain comprehensive information more quickly, with a trade-off between throughput and analytical capacity.

Recent research has placed episodic reinforcement learning(RL) alongside model-free and model-based RL on the list ofprocesses centrally involved in human reward-based learning.In the present work, we extend the unified account of model-free and model-based RL developed by Wang et al. (2017) tofurther integrate episodic learning. In this account, a genericmodel-free "meta-learner" learns to deploy and coordinate allof these RL algorithms. The meta-learner is trained on a broadset of novel tasks with limited exposure to each task, suchthat it learns to learn about new tasks. We show that whenequipped with an episodic memory system inspired by theoriesof reinstatement and gating, the meta-learner learns to use thesame pattern of episodic, model-free, and model-based RLobserved in humans in a task designed to dissociate among theinfluences of these learning algorithms. We discuss implicationsand predictions of the model.

Abstract Background Cardiovascular magnetic resonance T1ρ mapping may detect myocardial injuries without exogenous contrast agent. However, multiple co-registered acquisitions are required, and the lack of robust motion correction limits its clinical translation. We introduce a single breath-hold myocardial T1ρ mapping method that includes model-based non-rigid motion correction. Methods A single-shot electrocardiogram (ECG)-triggered balanced steady state free precession (bSSFP) 2D adiabatic T1ρ mapping sequence that collects five T1ρ-weighted (T1ρw) images with different spin lock times within a single breath-hold is proposed. To address the problem of residual respiratory motion, a unified optimization framework consisting of a joint T1ρ fitting and model-based non-rigid motion correction algorithm, insensitive to contrast change, was implemented inline for fast (~ 30 s) and direct visualization of T1ρ maps. The proposed reconstruction was optimized on an ex vivo human heart placed on a motion-controlled platform. The technique was then tested in 8 healthy subjects and validated in 30 patients with suspected myocardial injury on a 1.5T CMR scanner. The Dice similarity coefficient (DSC) and maximum perpendicular distance (MPD) were used to quantify motion and evaluate motion correction. The quality of T1ρ maps was scored. In patients, T1ρ mapping was compared to cine imaging, T2 mapping and conventional post-contrast 2D late gadolinium enhancement (LGE). T1ρ values were assessed in remote and injured areas, using LGE as reference. Results Despite breath holds, respiratory motion throughout T1ρw images was much larger in patients than in healthy subjects (5.1 ± 2.7 mm vs. 0.5 ± 0.4 mm, P

Background and aims: Experiencing acute stress is common in behavioral addictions such as gambling disorder. Additionally, like most substance-induced addictions, aberrant decision-making wherein a reactive habit-induced response (conceptualized as a Model-free [MF] in reinforcement learning) suppresses a flexible goal-directed response (conceptualized as a Model-based [MB]) is also common in gambling disorder. In the current study we investigated the influence of acute stress on the balance between habitual response and the goal-directed system. Methods: A sample of N 5 116 problem gamblers (PG) and healthy controls (HC) performed an acute stress task -- the Socially Evaluated Cold pressure task (SECPT) -- or a control task. Self-reported stress and salivary cortisol were collected as measures of acute stress. Following the SECPT, participants performed the Two-Step Markov Task to account for the relative contribution of MB and MF strategies. Additionally, verbal working memory and IQ measures were collected to account for their mediating effects on the orchestration between MB/MF and the impact of stress. Results: Both groups had comparable baseline and stress-induced cortisol response to the SECPT. Non-stressed PG displayed lower MB learning than HC. MANOVA and regression analyses showed a deleterious effect of stress-induced cortisol response on the orchestration between MB and MF learning in HC but not in PG. These effects remained when controlling for working memory and IQ. Discussion and Conclusions: We found an abnormal pattern of modulation of stress on the orchestration between MB and MF learning among PG. Several interpretations and future research directions are discussed. [ABSTRACT FROM AUTHOR]

Rotating machinery is fundamental in industry, gearboxes especially. However, failures may occur in their transmission components due to regular usage over long periods of time, even when operations are not intense. To avoid such failures, Structural Health Monitoring (SHM) techniques for damage prediction and in-advance detection can be applied. In this regard, correlations between measured signal variations and damage can be inspected using Artificial Intelligence (AI), which demands large numbers of data for training. Since obtaining signal samples of damaged components experimentally is currently unviable for complex systems due to destructive test costs, model-based numerical approaches are to be explored to solve this problem. To address this issue, this work applied an innovative hybrid Finite Element Method (FEM)–analytical approach, reducing computational effort and increasing performance with respect to traditional FEM. With this methodology, a system can be simulated with accuracy and without geometrical simplifications for healthy and damaged cases. Indeed, considering different positions and dimensions of damages (e.g., cracks) on the tooth roots of gears can offer new ways of damage investigation. As a reference to validate healthy systems and damage cases in terms of eigenfrequencies, a back-to-back test rig was used. Numerical simulations were performed for different cases, resulting in vibrational spectra for systems with no damage, with damage, and with damage of different intensities. The vibration spectra were used as data to train an Artificial Neural Network (ANN) to predict the machine state by Condition Monitoring (CM) and Fault Diagnosis (FD). For predicting the health and the intensity of damage to a system, classification and multi-class classification methods were implemented, respectively. Both sets of classification results presented good prediction agreement. [ABSTRACT FROM AUTHOR]

The continuous development of new materials and larger and/or more complex structures drives the need for the development of more robust, accurate, and sensitive Structural Health Monitoring (SHM) techniques. In the present work, a novel vibration-based damage-detection method that contributes into the SHM field is presented using Metaheuristic algorithms coupled with optimal Finite Element Models that can effectively localize damage. The proposed damage-detection framework can be applied in any kind of detailed structural FE model, while requiring only the output information of the dynamic response of the structure. It can effectively localize damage in a structure by highlighting not only the affected part of the structure but also the specific damaged area inside the part. First, the optimal FE model of the healthy structure is developed using appropriate FE model updating techniques and experimental vibration measurements, simulating the undamaged condition. Next, the main goal of the proposed method is to create a damaged FE model that approximates the dynamic response of the damaged structure. To achieve this, a parametric area is inserted into the FE model, changing stiffness and mass to simulate the effect of the physical damage. This area is controlled by the metaheuristic optimization algorithm, which is embedded in the proposed damage-detection framework. On this specific implementation of the framework, the Particle Swarm Optimization (PSO) algorithm is selected which has been used for a wide variety of optimization problems in the past. On the PSO's search space, two parameters control the stiffness and mass of the damaged area while additional location parameters control the exact position of the damaged area through the FE model. For effective damage localization, the Transmittance Functions from acceleration measurements are used which have been shown to be sensitive to structural damage while requiring output-only information. Finally, with proper selection of the objective function, the error that arises from modeling a physical damage with a linear damaged FE model can be minimized, thus creating a more accurate prediction for the damaged location. The effectiveness of the proposed SHM method is demonstrated via two illustrative examples: a simulated small-scale model of a laboratory-tested vehicle-like structure and a real experimental CFRP composite beam structure. In order to check the robustness of the proposed method, two small damage scenarios are examined for each validation model and combined with random excitations. [ABSTRACT FROM AUTHOR]

Complex robotic systems, such as humanoid robot hands, soft robots, and walking robots, pose a challenging control problem due to their high dimensionality and heavy non-linearities. Conventional model-based feedback controllers demonstrate robustness and stability but struggle to cope with the escalating system design and tuning complexity accompanying larger dimensions. In contrast, data-driven methods such as artificial neural networks excel at representing high-dimensional data but lack robustness, generalization, and real-time adaptiveness. In response to these challenges, researchers are directing their focus to biological paradigms, drawing inspiration from the remarkable control capabilities inherent in the human body. This has motivated the exploration of new control methods aimed at closely emulating the motor functions of the brain given the current insights in neuroscience. Recent investigation into these Brain-Inspired control techniques have yielded promising results, notably in tasks involving trajectory tracking and robot locomotion. This paper presents a comprehensive review of the foremost trends in biomimetic brain-inspired control methods to tackle the intricacies associated with controlling complex robotic systems., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Mompó Alepuz, Papageorgiou and Tolu.)

Experience replay is widely used in AI to bootstrap reinforcement learning (RL) by enabling an agent to remember and reuse past experiences. Classical techniques include shuffled-, reversed-ordered- and prioritized-memory buffers, which have different properties and advantages depending on the nature of the data and problem. Interestingly, recent computational neuroscience work has shown that these techniques are relevant to model hippocampal reactivations recorded during rodent navigation. Nevertheless, the brain mechanisms for orchestrating hippocampal replay are still unclear. In this paper, we present recent neurorobotics research aiming to endow a navigating robot with a neuro-inspired RL architecture (including different learning strategies, such as model-based (MB) and model-free (MF), and different replay techniques). We illustrate through a series of numerical simulations how the specificities of robotic experimentation (e.g., autonomous state decomposition by the robot, noisy perception, state transition uncertainty, non-stationarity) can shed new lights on which replay techniques turn out to be more efficient in different situations. Finally, we close the loop by raising new hypotheses for neuroscience from such robotic models of hippocampal replay. [ABSTRACT FROM AUTHOR]

Deep reinforcement learning (DRL) recently has attained remarkable results in various domains, including games, robotics, and recommender system. Nevertheless, an urgent problem in the practical application of DRL is fast adaptation. To this end, this article proposes a new and versatile metalearning approach called fast task adaptation via metalearning (FTAML), which leverages the strengths of the model-based methods and gradient-based metalearning methods for training the initial parameters of the model, such that the model is able to efficiently master unseen tasks with a little amount of data from the tasks. The proposed algorithm makes it possible to separate task optimization and task identification, specifically, the model-based learner helps to identify the pattern of a task, while the gradient-based metalearner is capable of consistently improving the performance with only a few gradient update steps through making use of the task embedding produced by the model-based learner. In addition, the choice of network for the model-based learner in the proposed method is also discussed, and the performance of networks with different depths is explored. Finally, the simulation results on reinforcement learning problems demonstrate that the proposed approach outperforms compared metalearning algorithms and delivers a new state-of-the-art performance on a variety of challenging control tasks. [ABSTRACT FROM AUTHOR]

Context: Testing is one of the techniques that can assure the quality of software including the domain of Software Product Line (SPL). Various techniques have been deliberated to enhance the quality of SPL including Model-based Testing (MBT). Objective: The objective of this study is to analyze and classify trends of MBT in SPL covering the solutions, issues and evaluation aspects by using taxonomy form. Method: A Systematic Literature Review (SLR) was conducted involving 63 primary studies from different sources. The selected studies were categorized based on their common characteristics. Results: Several findings can guide future research on MBT for SPL. The important finding is that the multiple measurements are still open to improving current metrics to evaluate test cases in MBT for SPL. The multiple types of measurement required a trade-off between maximization and minimization results to ensure the testing method which could satisfy multiple test criteria for example cost and effectiveness at the same time. [ABSTRACT FROM AUTHOR]

Experience replay is widely used in AI to bootstrap reinforcement learning (RL) by enabling an agent to remember and reuse past experiences. Classical techniques include shuffled-, reversed-ordered- and prioritized-memory buffers, which have different properties and advantages depending on the nature of the data and problem. Interestingly, recent computational neuroscience work has shown that these techniques are relevant to model hippocampal reactivations recorded during rodent navigation. Nevertheless, the brain mechanisms for orchestrating hippocampal replay are still unclear. In this paper, we present recent neurorobotics research aiming to endow a navigating robot with a neuro-inspired RL architecture (including different learning strategies, such as model-based (MB) and model-free (MF), and different replay techniques). We illustrate through a series of numerical simulations how the specificities of robotic experimentation (e.g., autonomous state decomposition by the robot, noisy perception, state transition uncertainty, non-stationarity) can shed new lights on which replay techniques turn out to be more efficient in different situations. Finally, we close the loop by raising new hypotheses for neuroscience from such robotic models of hippocampal replay.