Search

Visual SLAM algorithms can obtain a large amount of texture information from the environment and usually perform very well in static scenes, but there are a large number of irregular dynamic points when running in dynamic scenes, which can lead to increased error in SLAM feature point matching and loss of tracking localization. To address this challenge this paper proposes a SLAM system (YVG‐SLAM) that adapts to dynamic scenes. YVG‐SLAM is an improvement on the ORB‐SLAM3 algorithm, integrating the view geometry algorithm and the current powerful YOLOv5 algorithm, while proposing a dynamic feature removal strategy without a priori assumptions to reduce the influence of dynamic targets. In this paper, we first lighten the YOLOv5 algorithm, then process the image frame to get the bounding box, then use the view geometry algorithm to determine the dynamic feature points on the image frame, and finally remove the moving objects in the image frame according to the dynamic feature recognition strategy proposed in this paper. The performance of the algorithm is tested on the TUM RGB‐D and BONN RGB‐D datasets. The experimental results show that the robustness of the proposed algorithm on most video sequences is better than the existing SLAM algorithms dealing with dynamic scenes such as Detect‐SLAM and DynaSLAM, and the RMSE index of the absolute trajectory error measured on high‐speed dynamic video sequences can be significantly reduced by more than 93% compared to ORB‐SLAM3. © 2024 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

YOLOv7 is one of the most effective algorithms for one‐stage detectors. However, when it is applied to pedestrian detection tasks in the industrial scene, it is still challenging for complex environments and multi‐scale changes of pedestrians. This paper proposes a new pedestrian detector for the industrial scene based on improved YOLOv7‐tiny and named as GP‐YOLO. First, the neck of YOLOv7‐tiny is replaced by RepGFPN structure, make full use of multi‐scale features to enhance the detection accuracy of objects with large‐scale changes. Second, a new gnconv branch is added to the feature fusion module, and the high‐order spatial interaction capability is introduced to further enhance the target detection accuracy. Finally, a lightweight method based on PModule is proposed, on this basis, a PConv bottleneck is designed to reduce the FLOPs and enhance the feature extraction. Experiments on a self‐made Industrial Pedestrian Data set show that before lightweight, the proposed algorithm achieves a 3.2% improvement in mAP@0.5:0.95 and a 3.7% improvement in Recall compared to the baseline YOLOv7‐tiny. After lightweight GP‐YOLO, compared to non‐lightweight, parameters and FLOPs are decreased by 26% and 23%, respectively, the mAP@0.5:0.95 is decreased by only 1.1% and the Recall is decreased by only 1.3%, which remains at a high level. Compared with baseline YOLOv7‐tiny, the lightweight GP‐YOLO has similar parameters and FLOPs, but the mAP@0.5:0.95 is increased by 2.1%, and the Recall is increased by 2.4%. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Research on global harmonic mitigation against distributed and high‐proportioned nonlinear loads in modern distribution networks has received extensive attention. In this paper, a distributed compensation method for active power filters (APFs) is proposed, which utilizes a small number of APFs with adjustable reference to alleviate the global voltage distortion caused by a large number of distributed nonlinear loads. The model predictive control (MPC) is employed to calculate the optimal current reference for compensators at each optimization moment by predicting the system's performance in a certain future time period. Furthermore, the Kalman filter (KF) algorithm is used to estimate the initial harmonic voltage for the MPC, reducing the unfavorable effects of the model uncertainty and measurement noise. The test results based on a seven‐bus distribution network in MATLAB are provided to verify the effectiveness of the proposed method. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In this paper, a new dual mode blind equalization method using neighborhood symbol‐based decision and decision‐directed (NSBD‐DD) algorithm is proposed to equalize the satellite communication signals. The proposed algorithm adopts the combined cost function to employ the advantages of symbol decision algorithm and the decision‐directed algorithm, which can further improve the universality of the proposed algorithm. Compared with the conventional blind equalization algorithm, the compensation ability of intersymbol interference and convergence speed of the proposed algorithm can be improved simultaneously. To verify the performance of the proposed algorithm, the computer simulation is performed, and the simulation results show that the residual intersymbol interference of the 64QAM signals can be decreased to −50 dB under the satellite channel, which has 30 dB lower than multi‐modulus algorithm (MMA), 26 dB lower than NSBD algorithm, and 10 dB lower than SBD algorithm. And compared with SBD algorithm, the number of iterations required for the NSBD‐DD algorithm can be decreased from 8000 iterations to 6000 iterations, which convergence speed is increased by 25%. © 2023 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Obstructive sleep apnea (OSA) is a common sleep disorder and is diagnosed by polysomnography (PSG) as the gold standard. However, PSG is a time‐consuming and costly test, and patients have to endure long waits before receiving a PSG test in a hospital. In view of this, portable and wearable screening tools for OSA prediction have been developed recently as a low‐cost and easy‐to‐use method. In this paper, an OSA detection model, based on regression approach, using unsegmented electrocardiography (ECG) signals is developed to directly estimate the apnea‐hypopnea index (AHI) value, which is the first report in the literature. In this manner, this work can provide more details of OSA assessment to users and doctors. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Machine learning (ML) has become a rising sophisticated technological application trend in the electrical industry in recent years. Such innovation provides optional methodologies for many existing applications, such as power and load profile forecasting, reliability evaluation, substation behavior detection and state observation of electrical equipment, and so on. This paper presents a review of various supervised and unsupervised ML techniques and applications for electrical power systems, including generation, transmission, distribution and micro‐grid. The algorithms and applications are mainly summarized from IEEE journals and the interest of this paper shows the roles and developments of most used algorithms and its corresponding extensions and performance in different applications. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

This paper introduces the concept of intermittently connected sensor networks where sensors are sleep most of time and become active randomly and rarely. In the networks, packet collision probability and power consumption can be reduced. Meanwhile, the delivery delay increases with the use of the low activation values. In this paper, we design the energy‐efficient intermittently connected sensor networks, considering this trade‐off of activation control. We first provide the algorithm that computes the optimal packet forwarding rule achieving the minimum delivery delay, which can be calculated using the activation rates. We then propose the activation rate control methods, which control the activation rates of the sensors based on the density of the sensors and/or the number of wake‐ups of sensors having relay packets. By doing this, we can improve the delay performance, while suppressing the increase in the packet collision probability and the power consumption of sensors. Through simulation experiments, we demonstrate the effectiveness of our activation control methods in the intermittently connected sensor networks. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Accurate assessment of lightning protection performance of extra high voltage transmission lines is important for transmission line design and different lightning protection. Based on the electro‐geometrical model, this paper introduces the angle influence factor, and presents an algorithm for shielding failure trip‐out rate of angle tower, and the existing algorithm is expanded. By establishing the three‐dimensional geometric model of the angle tower, the electrical geometric relations of conductors and ground wires on both sides are analyzed, respectively, and the projection area algorithm for shielding failure is modified. To further verify the application value of the calculation algorithm, the change of the shielding failure trip‐out rate of the typical angle tower with the bending angle of transmission line and the tower height in a 500 kV transmission line is analyzed. The analysis results show that the trip‐out rate of the outside corner is higher than that of the inside one, and with the increase of the angle and the height of the tower, the trip‐out rate shows an upward trend. When the tension tower height is 71.5 m and the angle increases from 0° to 80°, the outside shielding failure trip‐out rate increased by 10.9% and the inside decreased by 9.3%. Compared with the classical electro‐geometrical model calculation, the difference of trip‐out rate is 4.3% when the bending angle is 80°. When the tangent tower height is 70 m and the angle increases from 0° to 40°, the shielding failure trip‐out rate on the outside increased by 5.9% while the inside decreased by 6.2%. The research results provide a simple algorithm for solving bending transmission line and give some reference for lightning protection design of EHV transmission lines. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

This paper presents a deep learning‐based face recognition device. It aims to instantly remind users of the identities of those who they met before in parties or conferences. When greeted, one, who is correctly identified, may frequently feel more respected than another who is mistaken for someone else in the first place. The golden rule suggests one to treat others with respect, and then others will treat the one with respect in return. Empirically, making others feel respected can definitely facilitate communication among people for social and particularly business purposes. For this sake, this paper employed a pair of smart glasses as a hardware platform into which a deep learning‐based multi‐task cascaded convolutional networks and a FaceNet algorithm were integrated. This move provided a high face recognition accuracy and a strong robustness. In the end, this smart device as intended recognized captured images and displayed the identities of matched contacts in a face image database. Hopefully, this proposal can be applied to a wide diversity of fields. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In order to construct an efficient and reliable global path for unmanned surface vehicles (USVs), this paper proposes a USV global path planning algorithm based on an optimized ant colony algorithm (OACA). The algorithm constructs a USV global path planning model by comprehensively considering energy consumption costs and turning control costs. In order to ensure the adaptability of the model to the task environment, the USV's motion characteristics are analyzed with full consideration of external interference, and on this basis, the characterization of energy consumption costs and turning control costs is completed. In order to obtain the global optimal solution of the model and accelerate the convergence speed of the model, in the initialization process of the ant colony algorithm, the initial pheromone is distributed unevenly by introducing the distance relationship among the intermediate node, the starting point, and ending point to improve the initial search efficiency. In the iterative process of searching for the optimal solution, enhancing the guidance effect of the current optimal solution on the offspring ants by introducing a weight factor to improve the update rule of pheromone, which accelerates the convergence speed of the algorithm. In the last two steps of the deadlock path, the number of lost ants is reduced by introducing a penalty factor to punish the pheromone which can guarantee the diversity of ants, and the algorithm's premature convergence is overcome. The simulation results prove the effectiveness of this algorithm. © 2022 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In view of the poor performance of the traditional adaptive filtering algorithm in identifying sparse systems under impulse noise, an affine projection maximum correntropy criterion with compound inverse proportional function (APMCCCIPF) algorithm is proposed in this paper, and the compound inverse proportional function approaching l0 norm is introduced into the traditional affine projection maximum correntropy criterion (APMCC) algorithm. The fixed step‐size APMCCCIPF algorithm cannot obtain faster convergence speed and lower steady‐state error at the same time, and the most effective method to solve this contradiction is to change the fixed step‐size in the algorithm to a variable step‐size. Further, a variable step‐size factor is constructed in the APMCCCIPF algorithm, and a novel variable step‐size APMCCCIPF algorithm is proposed. A modified Gaussian function, which can resist impulse noise interference, is applied to adjust the step‐size change. Besides, the convergence and complexity of the proposed algorithms are analyzed. In the simulation of sparse system identification and echo cancellation under impulsive noise, the proposed algorithms not only achieve a lower and faster convergence rate, but also obtain a lower steady‐state error. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

This paper presents a parameter estimation based method to diagnose winding deformation and turn‐to‐turn fault of power transformers. First, an estimation model of transformer parameters is built, in which five equations are taken in account including voltage loop equation, active loss equation, input impedance equation, no‐load current equation and no‐load loss equation. Then, particle swarm optimization (PSO) is used to solve the model, and the characteristics of estimation data under winding faults are analyzed. At last, based on the estimation data, random forest (RF) algorithm is employed to classify transformer states and realize fault diagnostic. The simulation results show that the proposed parameter estimation method has high precision and are not affected by the factors including load power factors and the type and degree of winding deformation, and the impact of load rates can be avoided; the fault diagnostic scheme based on RF is quite sensitive and effective. The proposed method eliminates the need of transformer outage, and compared with other methods based on parameter estimation, it can distinguish winding deformation and turn‐to‐turn fault. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Missing data is a prevailing problem in predictive analytics. In this paper, a winner‐take‐all (WTA) autoencoder‐based piecewise linear model is developed to solve the nonlinear regression problem under the missing value scenario, which consists of two parts: an overcomplete WTA autoencoder and a gated linear network. The overcomplete WTA autoencoder is a stacked denoising autoencoder (SDAE) designed to play two roles: (1) to estimate the missing values; (2) to realize a sophisticated partitioning by generating a broad set of binary gate control sequences. Besides, an iterative algorithm with renewed teacher signals is developed to train the SDAE. On the other hand, the gated linear network with the generated binary gate control sequences implements a flexible piecewise linear model for nonlinear regression. By composing a quasi‐linear kernel based on the gate control sequences, the piecewise linear model is then identified in the same way as a support vector regression. Experimental results have shown that our proposed hybrid model has a better performance than traditional models. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

A new image dehazing algorithm based on adaptive sky region is proposed in this paper, which shows good fidelity in sky region and satisfying visual effect in non‐sky region. For robust sky segmentation, we propose a rough‐to‐fine method that can make a balance between efficiency and accuracy. Considering distribution of haze is inconsistent, we divide the input image into three parts and calculate their atmospheric lights respectively. To solve the problem of invalid dark channel prior, we make an improvement for the transmission estimation. Finally, image fusion is taken as a post processing that can solve the problem of partial darkness and ensure a visual pleasing result. The experimental results for both synthetic and natural hazy images demonstrate that our algorithm performs comparable or even better results than the state‐of‐the‐art methods in terms of various measurement indexes, such as the PSNR, SSIM, and so forth. Besides, the proposed algorithm can be also applied in FPGA platform due to the optimized performance. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In this paper, a new method is proposed to identify the parameters of roll motion equation using the response data of ship in regular waves. The proposed method is based on the fact that if the estimated parameters are close to the true parameters, then, its response are also close to the true response. Therefore, the parameter identification is achieved via minimizing the mean square error between the true response and the estimated response. To effectively solving the parameter minimization problem, an improved imperialist competition algorithm (ICA), called opposition based learning Gaussian bare bone imperialist competition algorithm (OBL‐GBBICA) is proposed. The proposed OBL‐GBBICA integrates the opposition learning and Gaussian sampling technique into ICA to enhance its exploration ability and speed up its convergence. Experimental results show that the proposed method can accurately identify the parameters of roll motion equation. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

This paper presents a novel distributed method for dynamic economic dispatch (DED) problems with cubic fuel cost functions based on the dual alternating direction method of multipliers (D‐ADMM) and interior point method (IPM). This scheme enables us to combine the excellent parallelism of ADMM with the superior convergence properties of the IPM. The idea adopted by the proposed algorithm is that the outer‐loop uses ADMM decoupling, and the inner‐loop uses IPM to solve the cubic polynomial optimization subproblems. When the unit set is divided into multiple partitions, D‐ADMM also has the nice central processing unit (CPU) runtime as the well as the number of iterations. In addition, we use a Decoupling‐Decomposition‐Backtracking and parallel improved multiple centrality corrections decoupling interior point method (DDB‐PIMCCD) to solve the aforementioned larger subproblems more efficiently. Finally, the numerical results, in both serial and parallel modes, on a set of 22 DED cases with the range of units from 8 to 1000 units show that the proposed method is very promising for large scale distributed DED problems. Because it can keep the unit information about each subset in secret to suit the market environment and can obtain high‐quality solutions with reasonable time. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In both the research and engineering fields, missing data is a serious problem that cannot be overlooked. Therefore, available datasets with missing data are a challenge to be modeled by conventional global prediction models. In this paper, we propose a hybrid model consisting of an autoencoder and a gated linear network for solving the regression problem under missing value scenario. A sophisticated modeling and identifying algorithm is developed. First, an extended affinity propagation (AP) clustering algorithm is applied to obtain a self‐organized competitive net dividing the datasets into several clusters. Second, a multiple imputation tool with top p% winner‐take‐all denoising autoencoders (DAE) is introduced to realize better predictions of missing values, in which rough estimates of missing values by using the mean imputation and similarity method within the clusters are used as teacher signals of DAE. Finally, a gated linear network is designed to construct a piecewise linear regression model with interpolations in the exact same way as a support vector regression with a quasilinear kernel composed using the cluster information obtained in the AP clustering step. Based on the experiments of five datasets, our proposed method demonstrates its effectiveness and robustness compared with other traditional kernels and state‐of‐the‐art methods, even on datasets with a large percentage of missing values. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

The conventional artificial neural network (ANN) model is an effective way to detect harmonic signals. However, the solution of the model cannot converge when the initial value of the excitation function (IVEF) is not suitable, causing a large error in harmonic detection, especially in the case of power system noise. We improve the back propagation (BP) neural network (BNN) model by using an adjustable excitation function and the basic inertial algorithm in this paper. Based on this model and combined with the glowworm swarm optimization (GSO) algorithm, a harmonic/inter‐harmonic detection method is proposed. The network first uses an improved GSO algorithm with an adaptive step to optimize the IVEF of BNN. Then, BNN is trained at this initial value, so that amplitude, phase, and frequency of harmonics/inter‐harmonics can be obtained. The simulation data analyses show that that the method has good stability, convergence, strong anti‐noise ability, and ultrahigh detection accuracy. It can accurately separate the integer and non‐integer harmonics. Compared with the traditional ANN algorithm and Hanning‐FFT, the detection accuracy of this algorithm can be improved by 2–4 orders of magnitude. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Dissolved gas analysis is an important way to diagnose transformer faults. Compared with the method of establishing a single classifier based on artificial intelligence for diagnosis, ensemble learning (EL) can combine multiple classifiers to achieve stronger generalization ability and better diagnostic performance. But the traditional EL belongs to homogenous ensemble in which the base learners are based on the same algorithm, so this kind of EL method lacks the differences among the base learners, as well as systematic combination strategy. For this problem, in the paper the Stacking ensemble strategy is applied to fault diagnosis. Multilayer perceptron, k‐nearest neighbor, decision tree and support vector machine are used as component learners, and random forest algorithm is used as a combination strategy to establish a Stacking diagnosis model. In addition, homogenous ensemble methods are applied to the above four algorithms. In the method, the content of five characteristic gases are taken as the input characteristic parameters. Primary diagnostic results can be obtained with each base classifier. Then the meta‐learner random forest model organizes the base classifiers, and uses the primary diagnostic output as the input of the meta‐learner for secondary diagnosis to get the final diagnosis. The experimental results show that the ensemble of multiple heterogeneous component learners can enhance the generalization ability of the model, and the diagnostic accuracy is better than single classifier and the homogenous ensemble classifier. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

The maximum correntropy criterion (MCC) algorithm has popularly been used in suppressing impulsive noise. It mainly relies on the choice of step size and kernel width. With the study of step size, this paper proposes a novel algorithm that updates the step size based on constructing a variable step size function that uses the moving weighted average algorithm to keep performance stable and sigmoid function to speed the convergence rate. The proposed algorithm is robust against impulsive noises and generates a large step size to accelerate the convergence in iteration beginning, and system mutation makes the system identification performance more stable when the error is small. Based on the improved variable kernel width maximum entropy criterion (IVKW‐MCC) algorithm, which overcomes the shortcoming of how to choose a reliable kernel width in the MCC algorithm, simulation results are compared with other MCC algorithms under impulsive noise interference. Simulations in the system identification scenarios show that the proposed algorithm has better performance in improving the convergence speed over other algorithms. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Evolutionary algorithm (EA) is a generic term for optimization algorithms inspired by biological optimization processes in the natural world. Although EAs are widely applied to complex real‐world problems because they do not require mathematical expression of target problems, theoretical design methods of EAs have not been established. To solve this issue, this paper proposes an approach of designing EAs within an algorithm framework in which mathematical characteristics are derived, and it also presents a concrete framework for that purpose. The presented framework, Regulated Evolution Strategies (RES) provides a stability analysis result that contributes to designing algorithms with expected behavior. The RES framework has a high degree of freedom in designing algorithms, so that it is possible to incorporate various contrivances such as local improvement of samples and reduction of constraint violations in RES‐based algorithms while maintaining the stability analysis result. Numerical experiments prove that the RES framework has a capability for designing high‐performance EAs. © 2020 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

Recently, multiperson tracking across non‐overlapping camera views is gaining increasing attention in the video surveillance field due to its importance for public security. Most existing methods find a correspondence of person images individually across each pair of adjacent two cameras. However, in such a naive approach, redundant matching where one of the person images is selected for the matching pair for several times, often occurs. It leads to tracking failure. To overcome this problem, in this paper, by considering the matching problem as an instance of the "marriage problem" that is well known in Economics, we propose a simultaneous image matching method for multiple persons based on the stable marriage algorithm. We confirmed that the proposed method outperforms some of the existing state‐of‐the‐art methods on several well‐known public datasets. © 2020 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

In this paper, we analyze the properties of the superior solution set search (SSSS) problem and point out the structural similarities between it and the multiobjective optimization problem. Based on the analyzed properties, we propose a new Firefly algorithm (FA) based on superior relations, which can search for the superior solution set by using the user's quantitative desire level as a search strategy. By introducing the distance of the problem space into the moving mechanism, FA makes it possible to search for multiple local optimal solutions by dividing the solution set into multiple groups. By analyzing this property of FA, we clarify the affinity between FA and the SSSS problem. We subsequently analyze the SSSS problem and FA, and discuss their properties together in a similar problem setting. We propose an FA based on superior relations as a new optimization technique for the superior solution set problem based on these analyses. Numerical experiments are then conducted using the SSSS problem and demonstrate the usefulness of the proposed method by comparing the performance of the proposed method with the conventional FA. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc. [ABSTRACT FROM AUTHOR]

Differential evolution algorithms have become one of the most competitive evolutionary algorithms, among which a chaotic differential evolution (CJADE) is a recently proposed state‐of‐the‐art variant. But CJADE still suffers from the premature convergence problem. This paper further improves CJADE by innovatively incorporating a success‐intensity‐based roulette wheel selection method into it. The resultant algorithm called SCJADE shows its superiority over its peer in terms of solution quality and convergence speed on IEEE CEC2017 optimization test suit. © 2021 Institute of Electrical Engineers of Japan. Published by Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]