Search

Finite-control-set model predictive control (FCS-MPC) has achieved superiority in managing multiphase induction machines due to its quick dynamic response, control flexibility, and overall good performance. Its advantages, including simplicity, computational efficiency, compensation for system perturbations, and effective handling of multivariable problems, have made it a competitive alternative in various industrial applications. Nevertheless, FCS-MPC has some limitations. It is highly dependent on the accuracy of the predictive model’s parameters. Unfortunately, the estimation of magnetizing inductance, the most critical factor, especially in the weakening field region, has not been studied yet. Focusing on this gap, this article proposes a technique to estimate the magnetizing inductance in the field weakening zone for a six-phase asymmetrical induction machine driven by the FCS-MPC. Experimental tests have verified the effectiveness of the proposed method, considering stator currents and rotor speed tracking, as well as a reduction in the ( $x-y$ ) currents.

The increase in air traffic in the recent years has motivated the development of technologies to monitor air space and warn about possible collisions by predicting the trajectories that will be followed by aircraft. In this field, neural networks have become prominent thanks to their potential to learn to predict maneuvers without providing aspects that are difficult to model such as atmospheric conditions, or detailed aircraft parameters. A variety of models have been proposed; however, these are often tested in very limited setups, leaving many unanswered questions about how they perform in certain conditions, or whether or not their accuracy can be improved by training models for specific trajectories, using additional features, predicting more distant points directly, etc. This may be problematic for researchers or developers of these systems, who have no way of knowing what strategies will yield the best results. We have identified ten open research questions that have not been answered through in-depth testing. This motivated us to carry out a novel experimental study that performs aircraft trajectory prediction with several dozens configuration variants to answer the aforementioned questions by means of a much more complete evaluation. Some of the conclusions of our study stand in contrast with some popular practices in the state of the art, which casts some doubts on the simplicity of their application; for example, differential features are crucial for proper performance but are not mentioned by most studies, while complex, more elaborate models may lead to worse results than simple ones. Other important insights include the benefit from specialized models in more challenging scenarios, the influence of the known trajectory length in said scenarios, the step degradation of predictions when predicting further into the future, or the detrimental effect of adding additional features. These insights should help guide future research about the application of neural networks when it comes to aircraft trajectory prediction and their eventual inclusion in final systems.

The use of sensors to detect or measure ripening changes in fruit is a growing area of interest to the scientific community. Colorimeters are commonly employed for color and shade identification; however, their usage to measure the color parameters of fruit rinds based on image analysis can be expensive. This article presents a non-invasive and low-cost optoelectronic system for detecting color changes in oranges to predict the ripening stage. The system utilises a 1 W white LED as a light source, an RGB photodiode array, and two plastic optical fibres bundled in parallel to form the head of an extrinsic sensor. A microcontroller is employed for model integration and data acquisition. The evolution of the skin color of the fruit was monitored until over-ripeness was evident. The sensor was designed to detect the color changes; the CIE L*a*b* color difference between the optoelectronic device results and those obtained by colorimetry was 2.6–4.5. To predict the ideal conditions for fruit handling and determine the maturity level, a multilevel perceptron ANN was trained, achieving an accuracy of 96.4%. In addition, an overall precision of 96.6% was achieved when classifying fruit into three maturity categories (under-ripe, ripe, and over-ripe), and the error was 3.4%. The combination of the optoelectronic device and ANN improves considerably this fruit color classification accuracy, can facilitate the determination of the optimal time for consumption, and optimize the postharvest process efficiency.

This article addresses the challenge of attitude synchronization of a group of rigid bodies (considered to be agents) with zero final angular velocity. A collaborative control strategy is proposed to tackle this issue, using decentralized consensus within a leader-follower scheme, and the communication between agents is activated by events. The attitude is parameterized by the unit quaternion. The control law incorporates an event function specifying the moment when the $i$ -th agent should transmit attitude and angular velocity information to its neighbors. The communication topology between agents is modeled using a connected and undirected graph. The event-triggered communication produces asynchronous information exchange, reducing data traffic without compromising system performance. Furthermore, in practical scenarios such as networks of satellites or aerial robots, the proposed strategy could reduce the use of communication channel bandwidth. The Lyapunov method is utilized to analyze the stability of the overall system. Numerical simulation results confirm the proposal’s effectiveness.

Image segmentation is the process of partitioning an image into multiple regions or objects, each representing a coherent and meaningful part of the image. Segmentation methods are highly sensitive to the lack of homogeneity in regions or objects owing to noise and intensity inconsistencies. Under such conditions, most approaches exhibit poor quality performance. This paper proposes an agent-based model approach for homogenization of images to reduce the presence of noisy pixels and undesirable artifacts. In our approach, each pixel in the image represents an agent, and a set of rules evaluates the states of neighboring agents to modify the intensity values of each pixel iteratively until different regions from the image assume homogeneous grayscale levels. The proposed method has been used in combination with the Otsu’s method to evaluate its performance in image segmentation. The approach was evaluated with different types of images considering their homogeneity. Experimental results indicated that the proposed approach produces better-segmented images in terms of quality and robustness.

Adiabatic quantum-flux-parametron (AQFP) logic is an ultra-low-power superconductor logic family. AQFP logic gates are powered and clocked by dedicated clocking schemes using ac excitation currents to implement an energy-efficient switching process, adiabatic switching. We have proposed a low-latency clocking scheme, delay-line clocking, and demonstrated basic AQFP logic gates. In order to test more complex circuits, a serializer/deserializer (SerDes) should be incorporated into the AQFP circuit under test, since the number of input/output (I/O) cables is limited by equipment. Therefore, in the present study we propose and develop a novel SerDes for testing delay-line-clocked AQFP circuits by combining AQFP and rapid single-flux-quantum (RSFQ) logic families, which we refer to as the AQFP/RSFQ hybrid SerDes. The hybrid SerDes comprises RSFQ shift registers to facilitate the data storage during serial-to-parallel and parallel-to-serial conversion. Furthermore, all the component circuits in the hybrid SerDes are clocked by the identical excitation current to synchronize the AQFP and RSFQ parts. We fabricate and demonstrate a delay-line-clocked AQFP circuit (8-to-3 encoder, which is the largest delay-line-clocked circuit ever designed) integrated with the hybrid SerDes at 4.2 K up to 4.5 GHz. Our measurement results indicate that the hybrid SerDes enables the testing of delay-line-clocked AQFP circuits with only a few I/O cables and is thus a powerful tool for the development of very large-scale integration AQFP circuits.

Model Predictive Controller (MPC) is widely used as a technique for path tracking control since it allows for dealing with system constraints and future forecasts. However, the performance of MPC is directly affected by the adopted model. A complex dynamic model can guarantee accuracy in path tracking but may not be suitable in computational terms. On the other hand, a simplified model may not capture essential nonlinear aspects. Thus, to cope with these problems, this paper deals with data-driven tire modeling to improve autonomous ground vehicle path tracking control. The main contribution of the present work is to show that neural tires can capture the nonlinearities present in the interaction between lateral and longitudinal vehicle dynamics, with a reduced computational cost for predictive controllers. Simulated and experimental tire data are approximate to design data-driven tire models using radial basis function and multilayer perceptron neural networks. Then, based on ground vehicles with neural tires, model predictive controllers are designed to regulate wheel torque and steering angle inputs. Comparative tests were conducted to compare the proposed data-driven MPC approach with the classical nonlinear MPC controller. The results show that the neural tires approximate nonlinear tire models and experimental data with arbitrary precision in terms of accuracy and error-based metrics. The proposed methodology was successfully applied to perform trajectory and velocity tracking of ground vehicles. In addition, MPC with a neural tire model as prediction inference reduces the computational effort compared to traditional approaches.

In the last decade, model predictive control (MPC) has been widely studied in power converters, such as voltage source inverters (VSIs). Unfortunately, MPC often presents a high computational burden that limits their applicability, especially when driving multilevel inverters (MLIs) because of their higher number of switching combinations than two-level inverters. As a result, some strategies have been developed to reduce the computational complexity of MPC. One of the most relevant is the use of artificial neural networks (ANNs) to approximate the behavior of an MPC. However, ANNs require to be evaluated at bounded inputs. Otherwise, their response cannot be guaranteed to be a good approximation of the controller they learned from. Furthermore, when driving an LC-filtered VSI, the inductor current can present high peaks due to the cross-coupling effect between the inductor and the capacitor. These current peaks can cause physical damages and loss of performance of an ANN-emulated MPC. This paper presents a new constrained modulated MPC (M2PC), better suited for ANN emulation, to overcome these issues. The proposed strategy evaluates the cost function once per switching region, allowing easy and intuitive constraint inclusion. Additionally, an overmodulation stage is used to handle negative duty cycles and enhance disturbance rejection. Finally, the proposal is validated through simulations, using MATLAB-Simulink, taking into account different load conditions. Simulations show that the constrained M2PC keeps the inductor current below its desired limit while having a good performance (low harmonic distortions and fast dynamics) even when the inverter operates near its boundaries.

Early alert fire and smoke detection systems are crucial for daily and security management decision-making. Recent literature approaches are based on Deep Learning (DL) models. Efficient models are required for hardware-constrained systems, such as mobile devices, embedded systems, and robotics achieving high performance at low power consumption. For this research, we designed a novel specific-purpose model for fire and smoke recognition using still images and the study of state-of-the-art convolution techniques to improve the trade-off between accuracy and complexity. In this regard, the literature suggests that the inverted residual block, the depthwise and octave convolution techniques, reduces the model’s size and computation requirements working well by themselves. In this work, we propose the KutralNext architecture, an efficient model for single- and multi-label fire and smoke recognition tasks. Additionally, a more efficient architecture KutralNext+, demonstrates that those convolution techniques achieve a better trade-off combined, reaching an 84.36% average test accuracy in FireNet, FiSmo, and FiSmoA fire datasets. The KutralSmoke and FiSmo fire and smoke datasets attained an 81.53% average test accuracy. Furthermore, a previous fire and smoke recognition model considered, FireDetection, KutralNext uses 59% fewer parameters, and KutralNext+ requires 97% fewer flops and is 4x faster.

Gesture recognition is an ideal means of interaction because it allows users not to have to make contact with any surface, which is a safe and hygienic means, especially in the pandemic situation that is occurring worldwide. However, gesture recognition is not a new discipline and it has been researched for many years but this type of interaction has not succeeded in replacing the keyboard and mouse. It is very useful to know about the advances that are being made with artificial intelligence in gesture recognition to be able to perform a more robust and reliable gesture recognition with a low response time. As it is, deep learning is being integrated into various areas to increase improvement in performance and one such area is artificial intelligence. In this way, there is the possibility that in the future the recognition of gestures will be a viable option as a means of daily interaction for the user and the main objective of this paper is to contribute to that process. For this reason, this study has analyzed 571 papers related to gesture recognition and artificial intelligence. This analysis has extracted relevant information related to scientific production, such as the most productive authors and journals or the most pertinent articles on the subject. Furthermore, we have developed our own model, which shows the relationship between the types of gesture recognition and the artificial intelligence techniques that have been applied for this task.

A variable-speed predictive torque control (PTC) of a six-phase induction machine (SPIM) fed by a multi-modular matrix converter (M-MMC) is presented in this paper. This new system inherits the advantages of both multiphase machines (i.e., inherent fault-tolerance and better distribution of the power/current per phase compared to three-phase machines) and the M-MMC (i.e., smaller size and weight, and has no bulky storage elements). In addition, with the M-MMC topology, it is possible to use two power generation sources of different characteristics instead of a single three-phase source. This article proposes a method that reduces the number of calculations significantly compared to conventional PTC to overcome this issue. Simulation and experimental results are provided for speed control, module failure performance and tolerance to the variation of the magnetization inductance of the SPIM. An experimental test bench on a customized SiC-Mosfets based M-MMC and a 5.5 kW symmetrical SPIM has been used.

Shear Horizontal (SH) guided waves have been extensively used to estimate and detect defects in structures like plates and pipes. Depending on the frequency and plate thickness, more than one guided-wave mode propagates, which renders signal interpretation complicated due to mode mixing and complex behavior of each individual mode interacting with defects. This paper investigates the use of machine learning models to analyse the two lowest order SH guided modes, for quantitative size estimation and detection of corrosion-like defects in aluminium plates. The main contribution of the present work is to show that mode separation through machine learning improves the effectiveness of predictive models. Numerical simulations have been performed to generate time series for creating the estimators, while experimental data have been used to validate them. We show that a full mode separation scheme decreased the error rate of the final model by 30% and 67% in defect size estimation and detection respectively.

Nonlinear Model-based Predictive Control (NMPC) is a relevant research area having applications in the industrial sector. Traditionally, in this technique, gradient descent algorithms have been used to solve the related optimization problem. More recently, bio-inspired meta-heuristics have also been applied to this problem. However, only a few works have been devoted to testing solvers that use parameter control with self-adaptive traits, which allows mitigating the problem of offline parameter tuning in bio-inspired approaches. In this paper, we propose the novel Adaptive Modified Grey Wolf Optimization (AMGWO) and the Adaptive Moth-Flame Optimization (AMFO), for solving Nonlinear Model-based Predictive Control (NMPC) problems. To achieve this, a mechanism for individual leaders weighting and a crossover operator are introduced in AMGWO, and a simple self-adaptive parameter technique is applied in both meta-heuristics. The improved solvers are tested to perform the swing-up of a single inverted pendulum and attitude control of a satellite, which are nonlinear problems relevant for assessing control performance. Nonparametric statistical tests are applied to compare the improved meta-heuristics optimization outcomes with other five meta-heuristics, which shows that the self-adaptive parameter technique can significantly improve the performance when applied as an NMPC solver, as the AMFO and AMGWO statistically outperform or performs as well as all algorithms compared in both the pendulum and satellite control, respectively. This is important as improving the optimizer efficiency will lead to more accurate control and enable rapid hardware implementation.

Modern outdoor self-localizing computer vision applications require descriptors more than repeatability. The descriptors need to be invariant to light conditions and transformation changes to give support for efficient classification. This paper investigates a new framework based on the genetic algorithm to create and optimize extensible modular descriptors for specific outdoor environments. The algorithm returns descriptors with improved efficiency and classification performance. It controls the image processing and machine learning parameters and optimizes the descriptor size by activating the necessary modules. To show the strength of the descriptor, we compared it with the most commonly used standard descriptors on speed, accuracy, and invariance to light conditions, image resolution changes, scale, affine transformation, rotation, and classification. The results show that it has an average result in transformation invariance, and its description ability in sparse areas is significantly better than that of the most-used descriptors. The descriptor was also integrated into an augmented reality algorithm to create a self-regulating segmentation application.

We report the experimental observation of the polarization modulation instability (PMI) effect in all-normal dispersion (ANDi) microstructured optical fibers (MOFs) with quasi-continuous pumping. The small unintentional birefringence (~10-5), that any realistic non-polarization maintaining MOF exhibits, contributes to this nonlinear effect. PMI can produce two sidebands whose polarization state is orthogonal to the polarization of the pump. In this work, only one type of PMI process is observed, i.e., when the pump is polarized along the slow axis of the fiber and sidebands are generated in the fast axis mode. This PMI process was studied experimentally in two ANDi fibers with different dispersion features and pumped with long (700 ps) pump pulses at 1064 nm. Experimental results are compared with theoretical calculations, with reasonably good agreement.

The present manuscript experimentally demonstrates multi-wavelength generation from a thulium-doped all-fiber laser by implementing a multi-cavity Fabry-Perot filter. The filter, with a composition that relies of several silica-air and air-silica interfaces, provide a spectral reflection response that allows the laser to be operated in single-, dual- or triple-wavelength emissions by simple and clear adjustment of the polarization state. In this way, the fiber laser is capable to operate over a broad range of wavelengths extending from 1844 to 1928 nm. In addition, dual-wavelength operations are also feasible in the 1844 and 1928 nm regions, exhibiting a mode spacing of 4.41 nm and 1.62 nm, respectively. Beside this, the laser is able to switch to a sextuple-wavelength operation regime in the 1885 nm region, maintaining a minimum wavelength spacing of 1.53 nm. The laser demonstrates high wavelength stability, with variations below 60 pm at all analyzed wavelengths, and a minimum peak power fluctuation of ±2.77 dB.

Deciding on the continuous treatment of chronic diseases is vital in terms of economy, quality of life, and time. We present a holistic data mining approach that addresses the prediction of the therapeutic response in a panoramic and feedback way while unveiling relevant medical factors. Panoramic prediction makes it possible to decide whether the treatment will be beneficial without using previous knowledge and without involving unnecessary treatments. Feedback prediction can be more accurate prediction since it considers the results of previous stages of the treatment. A novel label encoding called simulated annealing and rounding (SAR) encoding is also proposed to help improve the accuracy of prediction in both approaches. To unveil the medical factors that make the treatment effective for patients, various techniques are applied to the prediction models found through the proposed approaches. Finally, this methodology is applied in the realistic scenario of analyzing electronic medical records of migraineurs under BoNT-A treatment. The results show a significant improvement in accuracy due to the use of SAR encoding, from close to 60% (baseline) to 75% with panoramic prediction, and up to around 90% when using feedback prediction. Furthermore, the following factors have been found to be relevant when predicting the migraine treatment responses: migraine time evolution, unilateral pain, analgesic abuse, headache days, and the retroocular component. According to doctors, these factors are also medically relevant and in alignment with the medical literature.

In image filtering, the classical lexicographical ordering is a popular method that cannot be directly applied for ordering colors in RGB color images. This is due to the fact that each color has similar importance and no order can be defined trivially a priori. In this work we propose an adaptive color lexicographical ordering framework for RGB color images where a color pixel is transformed into a real number. This transformation is weighted by statistical parameters from each color component histogram and used as the main component for color comparison. This approach seeks to avoid the arbitrariness since the order of the color component priorities is defined by the information extracted from the image itself. The proposed approach was tested by applying a median filter to reduce noise and a morphological approach to local contrast enhancement. In noise reduction, we compare our method with classical ordering techniques on images with different noise levels. Results show that our proposal outperformed the state-of-the art methods according to the Mean Absolute Error (MAE) measure, especially in those scenarios with higher noise levels. In contrast enhancement, the proposed framework outperformed the classical lexicographical ordering method according to Contrast Improvement Ratio (CIR) metric, especially when increasing the contrast factor. Our proposal generates less distortion than the state-of-the art methods ordering.

In this paper, a domestic refrigerator was tested to determine the power consumption for different positions of the shelves. The main contribution of this paper is the application of several strategies to analyze and model the power consumption of a domestic refrigerator when the fresh food shelves are changed of position. First, computer simulations were performed to analyze the power consumption using the moving average, the numerical derivative, and the Fourier transform. These simulations were used to study changes, periodic behaviors, minimum, maximum, and average values for the power consumption. Second, prediction methods were used to model the power consumption, these methods include cubic splines, the bilinear method, and artificial neural networks. With these methods, 2-D color graphs were built to predict the power consumption for any shelves positions. The validation results revealed that the cubic spline method provides the best results. Finally, it can be concluded that the models proposed in this paper provides new insights that can be used to design the internal compartments of the fresh food in order to try reducing the energy consumption.

Energy efficiency in cellular networks has gained great relevance due to the increasing power supply demands in new generation heterogeneous network (HetNet). On the other hand, interference coordination is a key aspect in HetNets resource management which directly affects the performance and energy consumption. Although these two aspect are intertwined, they have been studied separately so far. In this paper, we address the joint problem of energy saving and interference coordination in HetNets. We formulate the problem as a finite horizon Markov decision process (MDP) leveraging two facts: the user traffic demands usually follow periodic patterns, and the knowledge and prediction of network load is crucial in order to select efficient network configurations. Quality of Service (QoS) in the network is defined as the ratio of users meeting a requirement specified by the operator, and allows us to account for a minimum QoS requirement by including a constraint in the formulation of the MDP. To address this MDP, we propose an approximate dynamic programming (ADP) algorithm which selects energy efficient farsighted configurations with QoS guarantees achieving near optimal performance. This ADP algorithm is built upon: 1) the certainty equivalent control principle, which simplifies the complexity of the MDP, and 2) machine learning techniques: a neural network and a polynomial regressor which allow us to predict the QoS and the consumption of the network in advance. We evaluate our proposal in a LTE-A network simulator following the 3GPP guidelines, and the results obtained show that a joint control of the energy saving and interference coordination mechanisms results in a notably performance improvement compared to a disjoint control, in terms of both energy savings and QoS guarantees. Moreover, our proposal has the advantage of being adaptable to the operator QoS requirements.

Major challenges anticipated in the future C4ISR (command, control, communications, computers, intelligence, surveillance, and reconnaissance) operations involve rapid mission planning/ re-planning in highly dynamic, asymmetric, unpredictable, and network-centric environments. Developing decision support for such complex mission environments requires automated processing, interpretation, and development of proactive decisions using large volumes of structured, unstructured, and semi-structured data, while simultaneously decreasing the time necessary to arrive at a decision. To overcome this data deluge, there is a need for mastering information dominance via acquisition, fusion, and transfer of the right data/information/knowledge from the right sources in the right mission context to the right decision-maker (DM) at the right time for the right purpose (6R). The fundamental challenge in achieving the 6R is to conceive a generic framework that encompasses the dynamics of relevant contextual elements, their interdependence and correlation to the current and evolving situation, while taking into account the cognitive status of the DM. In this paper, we propose a context-driven proactive decision support (PDS) framework that comprises: 1) adaptive model-based dynamic graph models (e.g., Dynamic Hierarchical Bayesian Networks) and the concomitant inference algorithms for context representation, inference, and forecasting, 2) information selection, valuation, and prioritization methods for context-driven operations, including uncertainty management approaches, and 3) application of PDS concepts for courses of action recommendations across representative maritime operations.