Convergence Step Research Articles

An Enhanced Deep Q Network Algorithm for Localized Obstacle Avoidance in Indoor Robot Path Planning

Path planning is a key task in mobile robots, and the application of Deep Q Network (DQN) algorithm for mobile robot path planning has become a hotspot and challenge in current research. In order to solve the obstacle avoidance limitations faced by the DQN algorithm in indoor robot path planning, this paper proposes a solution based on an improved DQN algorithm. In view of the low learning efficiency of the DQN algorithm, the Duel DQN structure is introduced to enhance the performance and combined with a Prioritized Experience Replay (PER) mechanism to ensure the stability of the robot during the learning process. In addition, the idea of Munchausen Deep Q Network (M-DQN) is incorporated to guide the robot to learn the optimal policy more effectively. Based on the above improvements, the PER-D2MQN algorithm is proposed in this paper. In order to validate the effectiveness of the proposed algorithm, we conducted multidimensional simulation comparison experiments of the PER-D2MQN algorithm with DQN, Duel DQN, and the existing methodology PMR-DQN in the Gazebo simulation environment and examined the cumulative and average rewards for reaching the goal point, the number of convergent execution steps, and the time consumed by the robot in reaching the goal point. The simulation results show that the PER-D2MQN algorithm obtains the highest reward in both static and complex environments, exhibits the best convergence, and finds the goal point with the lowest average number of steps and the shortest elapsed time.

Fourth-order phase field modelling of brittle fracture with strong form meshless method

This study aims to find a solution for crack propagation in 2D brittle elastic material using the local radial basis function collocation method. The staggered solution of the fourth-order phase field and mechanical model is structured with polyharmonic spline shape functions augmented with polynomials. Two benchmark tests are carried out to assess the performance of the method. First, a non-cracked square plate problem is solved under tensile loading to validate the implementation by comparing the numerical and analytical solutions. The analysis shows that the iterative process converges even with a large loading step, whereas the non-iterative process requires smaller steps for convergence to the analytical solution. In the second case, a single-edge cracked square plate subjected to tensile loading is solved, and the results show a good agreement with the reference solution. The effects of the incremental loading, length scale parameter, and mesh convergence for regular and scattered nodes are demonstrated. This study presents a pioneering attempt to solve the phase field crack propagation using a strong-form meshless method. The results underline the essential role of the represented method for an accurate and efficient solution to crack propagation. It also provides valuable insights for future research towards more sophisticated material models.

Performance of PSO-GWO, VAGWO, and GWO optimization techniques for economic dispatch problem: studied case of the southeast algerian electrical network

Economic dispatch (ED) aims to identify the most cost-effective strategy for allocating power generation while meeting demand and adhering to the physical constraints of the power system. In this paper, three algorithms are proposed to solve the ED problem in power systems, including a hybrid Particle Swarm Optimization with Grey Wolf Optimizer (PSO-GWO), modified Velocity Aided Grey Wolf Optimizer (VAGWO), and Grey Wolf Optimizer (GWO). PSO is a meta-heuristic optimization technique designed to find the optimal solution to a problem by guiding the movement of particles within a defined exploration space. GWO is also a meta-heuristic optimization algorithm based on the natural behavior of grey wolves. In this paper, to enhance the performance of GWO, it is first combined with the PSO method. VAGWO is employed to refine the GWO formulation by optimizing the convergence steps. At the beginning of the iterations, the distance between the solutions (wolves) is increased to promote exploration of the search space. As the iterations progress, the step size is reduced, allowing for a more precise and efficient convergence toward the optimal solution. These algorithms will be implemented in Matlab software to minimize the fuel cost production of the following test networks: IEEE 30-bus network, Algerian 114-bus network, and Southeast Algerian network. These methods show that the PSO-GWO algorithm offers better results when compared through VAGWO and GWO.

Aerodynamic Parameter Identification Based on Multi Group Adaptive Search Optimization Algorithm

Abstract Due to the strong coupling and nonlinearity of spinning-stabilized correction projectile’s motion and dynamic model, its aerodynamic parameter identification can be transformed into a high-dimensional multi extremum optimization problem. To avoid the excessive computational complexity and susceptibility to local optima, this paper proposes a multiple swarm adaptive search optimization method. This method automatically divides the samples into several subpopulations through an adaptive density peak clustering method, and then searches and updates in each subpopulation. To balance convergence speed and population diversity, the update operator for each generation combines the random search and optimal convergence step size, whose weights can be adaptively adjusted based on the update rate and dispersion characteristics of the subpopulations. Applying this method to the aerodynamic parameter identification, and comparing with improved particle swarm optimization and genetic algorithm, simulation results show that a better identification result can be obtained at the same iterations.

A step function based recursion method for 0/1 deep neural networks

The deep neural network with step function activation (0/1 DNNs) is a fundamental composite model in deep learning which has high efficiency and robustness to outliers. However, due to the discontinuity and lacking subgradient information of the 0/1 DNNs model, prior researches are largely focused on designing continuous functions to approximate the step activation and developing continuous optimization methods. In this paper, by introducing two sets of network node variables into the 0/1 DNNs and by exploring the composite structure of the resulted model, the 0/1 DNNs is decomposed into a unary optimization model associated with the step function and three derivational optimization subproblems associated with the other variables. For the unary optimization model and two derivational optimization subproblems, we present a closed form solution, and for the third derivational optimization subproblem, we propose an efficient proximal method. Based on this, a globally convergent step function based recursion method for the 0/1 DNNs is developed. The efficiency and performance of the proposed algorithm are validated via theoretical analysis as well as some illustrative numerical examples on classifying MNIST, FashionMNIST and Cifar10 datasets.

An Optimal Strategy for Estimating Weibull distribution Parameters by Using Maximum Likelihood Method

Several methods have been used to estimate the Weibull parameters such as least square method (LSM), weighted least square method (WLSM), method of moments (MOM), and maximum likelihood (MLE). The maximum likelihood method is the most popular method (MLE). Newton-Raphson method has been applied to solve the normal equations of MLE’s in order to estimate the Weibull parameters. The method was used to find the optimal values of the Weibull distribution parameters for which the log-likelihood function is maximized. We tried to find the approximation solution to the normal equations of the MLE’s because there is no close form for get analytical solution. In this work, we tried to carry out a study that show the difference between two strategies to solve the MLE equations using Newton-Raphson algorithm. Both two strategies are provided an optimal solution to estimate the Weibull distribution parameters but which one more easer and which one converges faster. Therefore, we applied both strategies to estimate the Weibull’s shape and scale parameters using two different types of data (Real and simulation). We compared between the results that we got by applying the two strategies. Two studies have been done for comparing and selecting the optimal strategy to estimate Weibull distribution parameters using maximum likelihood method. We used some measurements to compare between the results such as number of steps for convergence (convergence condition), the estimated values for AIC, BIC and the RMSE value. The results show the numerical solution that we got by applying first strategy convergence faster than the solution that we got by applying second strategy. Moreover, the MRSE estimated by applying the first strategy is lower than the MRSE estimated by applying second strategy for the simulation study with different noise levels and different samples size.

Intrusion Detection in Wireless Sensor Networks Based on IPSO-SVM Algorithm

To optimize node energy consumption and improve its security, this paper uses the DEEC algorithm to layer WSN and reduce the probability of channel information collision and uses the weighted probability of cluster head election to optimize node energy expenditure, so that WSN can obtain a longer lifecycle. Improved Particle Swarm Optimization-based Support Vector Machine (IPSO-SVM) algorithm is used for intrusion detection and experimental testing in WSN. The results showed that the IPSO-SVM algorithm exhibited good convergence, with a convergence step size of 5 steps, which converged earlier than the Support Vector Machine Algorithm based on Particle Swarm Optimization (PSO-SVM), which had a convergence step size of 10 steps. The IPSO-SVM algorithm performed best in WSN intrusion detection, with the highest detection rate of 96.20% in Probe attack data detection, which was 0.80% higher than the Support Vector Machine Algorithm based on Genetic Algorithm (GA-SVM). The PSO-SVM algorithm had the lowest detection rate of 95.20%. The IPSO-SVM algorithm had a minimum false positive rate of 1.54% in Dos attack data detection. In terms of average training time, the IPSO-SVM algorithm had a minimum average training time of 323.45 seconds. Compared to the Low Energy Adaptive Clustering Hierarchy (LEACH) algorithm, the Distributed Energy Efficient Clustering (DEEC) algorithm performs better, has less energy consumption, and retains more nodes. The method adopted in this study can make WSN have a longer life cycle and ensure its security.

A guaranteed fixed-step convergence approach to discrete sliding mode control of linear disturbed systems

Current methods for achieving fixed-time convergence in continuous sliding mode control are extensively studied, whereas there are limited studies on fixed-step convergence in discrete sliding mode control. The control objective of this paper aims to develop control schemes assuring that the number of convergence steps of the tracking error remains within the desired range, even as the initial value increases, and this paper proposes a guaranteed fixed-step convergence approach to form the reaching law. An upper bound on convergent steps, independent of the initial value, is established, thereby having the fixed-step convergence property. The performance assessment, including attractiveness, invariance, and convergence steps, is provided to meet the required specifications. The steady-state band results offer guidance in selecting parameters to achieve the control objective. To achieve fixed-step convergence of the tracking error, one can develop a dead-beat terminal sliding mode control scheme, where the tracking error is governed by the prescribed error dynamics. To illustrate the approach more concretely, specific designs are proposed for both switching and non-switching reaching laws aimed at ensuring fixed-step convergence. The simulation and experiment results validate the performance evaluation and demonstrate effectiveness of the proposed control schemes.

Designing an evaluation tool for evaluating training programs of medical students in clinical skill training center from consumers’ perspective

IntroductionThe Clinical Skill Training Center (CSTC) is the first environment where third year medical students learn clinical skills after passing basic science. Consumer- based evaluation is one of the ways to improve this center with the consumer. This study was conducted with the aim of preparing a consumer-oriented evaluation tool for CSTC among medical students.MethodThe study was mixed method. The first phase was qualitative and for providing an evaluation tool. The second phase was for evaluating the tool. At the first phase, after literature review in the Divergent phase, a complete list of problems in the field of CSTC in medicine schools was prepared. In the convergent step, the prepared list was compared with the standards of clinical education and values of scriven. In the second phase it was evaluated by the scientific and authority committee. Validity has been measured by determining CVR and CVI: Index. The face and content validity of the tool was obtained through the approval of a group of specialists.ResultsThe findings of the research were in the form of 4 questionnaires: clinical instructors, pre-clinical medical students, and interns. All items were designed as a 5-point Likert. The main areas of evaluation included the objectives and content of training courses, implementation of operations, facilities and equipment, and the environment and indoor space. In order to examine the long-term effects, a special evaluation form was designed for intern.ConclusionThe tool for consumer evaluation was designed with good reliability and trustworthiness and suitable for use in the CSTC, and its use can improve the effectiveness of clinical education activities.

1 Sex Differences in Middle Cerebral Artery Velocity in Patients with Persistent Post-Concussion Convergence Insufficiency (PPCS-CI)

OBJECTIVES/GOALS: The objective of this study is to investigate the neurophysiological mechanism of vision therapy in male and female adolescents with persistent post-concussion convergence insufficiency (PPCS-CI). This study may improve diagnostics and inform more effective personalized point of care treatment strategies to remediate symptoms. METHODS/STUDY POPULATION: Participants (ages 11-25) were diagnosed with PPCS by a physician, CI was diagnosed by an optometrist and OBVAT was performed by certified therapists. Patients with PPCS-CI were randomly assigned to a therapy type (immediate therapy or natural recovery). Hemodynamic measures were examined in patients with PPCS-CI at baseline (1-3 months post-concussion), and post OBVAT to evaluate recovery outcomes. Non-invasive techniques were used to measure middle cerebral artery velocity (MCAv), blood pressure, heart rate, and end-tidal CO2 at rest and during objective symmetrical convergence step eye movements. Functional magnetic resonance imaging (fMRI) was acquired during convergence step eye movement experiments contrasted to sustained fixation and resting state data collection. RESULTS/ANTICIPATED RESULTS: To investigate the neural mechanism of OBVAT, eye movements, fMRI and physiological measures were collected in 8 patients with PPCS-CI (4 men and 4 women). Results show an 10% decrease in the MCA during 4-degree symmetrical convergent eye movement responses in males post-OBVAT and a 19% increase in MCA during convergent eye movement responses in females. Furthermore, there was a group level activation of the frontal eye fields, which improves post-OBVAT. The beta weights in the left frontal eye fields show a trending decrease in male patients post-OBVAT and trending increase in females. Males had a decrease in MCAv post OBVAT (baseline 83.6 ± 7.5 cm/sec & 75.7 ± 12.5 cm/sec post-OBVAT), while females show a significant increase post-OBVAT (baseline 53.77 ± 5.2 cm/sec & post-OBVAT 65.13 cm/sec ± 12.5). DISCUSSION/SIGNIFICANCE: This initial pilot demonstrates there may be different underlying pathophysiological outcomes associated with a concussion dependent on sex. This work may have direct implications on treatment strategies for male and female adolescent patients with PPCS-CI.

Optimization of artificial intelligence cloud computing in information management design

In order to promote the construction of enterprise informatization, the author studied the adaptive optimization method of Efficient Management Information System (EMIS). The author proposes to combine the fuzzy C-means algorithm to form a server clustering algorithm, and adds an improved Drosophila optimization algorithm to overcome the problems of slow Rate of convergence of GRNN and easy to fall into the minimum, and the cloud platform collects 23 performance indicators, the output results of the coordinated evolutionary algorithm are analyzed by the neighborhood rough set analysis of algorithms to select features to avoid the curse of dimensionality problem. The experimental results indicate that, compared with existing research results, the algorithm proposed by the author has increased its speed by 1.43, 3.22, and 3.72 times, respectively; In terms of convergence steps, they have also been reduced by 1.61, 5, and 6 times respectively, and when running the algorithm, the computer’s memory and CPU usage are controlled at around 50%, without affecting normal functionality. This proves that the task scheduling of the cloud platform is more balanced, and indirectly proves the accuracy of the algorithm’s clustering effect.

Time-averaged tomographic absorption spectroscopy for H2O diffusion in turbulent jet flow at room temperature

Tomographic absorption spectroscopy (TAS) is important for the combustion diagnostics of hydrogen and hydrocarbon fuels in various flow fields. In this study, we proposed and conducted time-averaged TAS for the reconstructions of mean H2O mole fraction XH2O in a dry, axisymmetric, turbulent jet flow with a Reynolds number of 5670 to reveal the effect of turbulence on absorption spectroscopy measurement. The absorption line of H2O near 1368.598 nm was selected and scanned at 1 kHz. The effect of turbulence on TAS, which is positively related to the turbulent intensity along the optical path, has been confirmed and can be effectively alleviated by conducting time averaging. It is shown that the proper selections of scan rate and the number of time steps for statistical convergence are required for time-averaged TAS. With time averaging and smoothing operations on the integrated absorbance, the two-dimensional mean XH2O distributions were reconstructed with the present method and the result showed a good agreement with CFD simulation. The mean deviation of the measured results can be reduced by over 25% when the number of time steps for average is 30000. The present study highlights the potential of TAS for the visualization and diagnostics of complex flows such as turbulent non-reactive flow and turbulent combustion.

Improving the stability of phase retrieval algorithm in speckle autocorrelation imaging by Huber penalty sparsity constraint

One of the most widely used methods for reconstructing images through a scattering medium is the Hybrid Input Output (HIO) phase retrieval algorithm. However, the stability and noise robustness of the HIO method are poor, as they are affected by initial values and noise during the imaging process. This paper presents an optimized HIO phase retrieval algorithm that utilizes the Huber penalty as a sparsity constraint to effectively enhance the quality and stability of the reconstruction algorithm. The introduced constraint is designed to guide the iteration process away from local optimal solutions, ensuring superior recovery images even with random initial values. Additionally, an adaptive step size parameter is incorporated to control the iterations and convergence steps, resulting in stronger robustness against noise effects. Moreover, the error criterion for image restoration is defined as the sum of the Fourier transform amplitude of the gradient. Experimental results demonstrate that the proposed phase retrieval algorithm achieves superior quality in the recovered images, while significantly improving noise immunity and stability for randomly selected initial values.

Automatic classification method of power user’s requirements text based on parallel naive Bayesian algorithm

In order to ensure the efficiency of power user’s requirements processing, an automatic classification method for demand test of power users based on parallel naive Bayesian algorithm is proposed. Polynomial naive Bayes is selected to build Hadoop cluster, and the feature words of power user’s requirements are selected through chi square test. The weight of each feature item is calculated by word frequency-inverse text frequency index method, and the weight sum of each category is calculated. The weight sum is input into naive Bayes algorithm to output the text classification results of power user’s requirements. At the same time, The naive Bayes classification algorithm is parallelized and encapsulated to reduce the cost of data movement and exchange in the classification process, and improve the operation efficiency of demand text classification of power user. The experimental results show that this method can accurately extract the feature words of power user’s requirements, effectively realize the automatic classification of power user’s requirements text, and have a more accurate classification effect. The average fitness value of the proposed method tends to be stable after more than 20 training times, and the number of network convergence steps is 7. When the ratio of energy function is about 0.4 and 0.6, the average IU value is the highest. When the required number of texts ranges from 500 to 1500, the delay time of text classification is 0.02 s, and the peak signal-to-noise ratio is more than 33, among which the highest peak signal-to-noise ratio is 42.52, and the normalization coefficient is 1.

A 4‐Arm PEG‐Thiodepsipeptide Precursor Enables Gelatinase‐Promoted Hydrogel Formation

AbstractIn situ hydrogelation of injectable precursors upon biological stimulus is relevant to generate hydrogels under mild conditions and, potentially, at a biological side of interest. Here, it is shown that hydrolytic enzymes can be used to initiate the formation of covalent hydrogel networks, realizing a cleavage‐leading‐to‐gelation strategy. For this purpose, a two‐component system is used, consisting of a 4‐arm polyethylene glycol‐thiodepsipeptide conjugate, PEG4TDPo containing the matrix metalloproteinases MMP‐2‐ and MMP‐9‐cleavable Ac‐Pro‐Leu‐Gly#SLeu‐Leu‐Gly‐ thiodepsipeptide sequence releasing a thiol upon hydrolysis, and a maleimide functionalized 4‐armed PEG (PEG4MAL). PEG4TDPo is synthesized in a PEG‐functionalization protocol involving convergent and divergent synthetic steps without the need for rigorous purification procedures. In a fluorometric assay, it is shown that the construct is in fact cleaved by both investigated MMPs. PEG4TDPo in the presence of 10 wt.% PEG4MAL formed hydrogels upon addition of MMP‐2 or ‐9 with average gelation times of 28 and 40 min, respectively, as is investigated by rheology. The much faster gelation times compared to the enzyme‐free system showed the specific input of the enzymatic reactions. The MMP‐assisted activation and crosslinking strategy can potentially become useful by targeting tissues showing an increased expression of MMPs, such as cancers, or to detect MMPs.

Learning-Assisted Algorithm Unrolling for Online Optimization with Budget Constraints

Online optimization with multiple budget constraints is challenging since the online decisions over a short time horizon are coupled together by strict inventory constraints. The existing manually-designed algorithms cannot achieve satisfactory average performance for this setting because they often need a large number of time steps for convergence and/or may violate the inventory constraints. In this paper, we propose a new machine learning (ML) assisted unrolling approach, called LAAU (Learning-Assisted Algorithm Unrolling), which unrolls the agent’s online decision pipeline and leverages an ML model for updating the Lagrangian multiplier online. For efficient training via backpropagation, we derive gradients of the decision pipeline over time. We also provide the average cost bounds for two cases when training data is available offline and collected online, respectively. Finally, we present numerical results to highlight that LAAU can outperform the existing baselines.

Selection of the convergence step of the Fx-LMS algorithm

The choice of a suitable convergence step that allows the fast convergence of the filtered-x LMS algorithm in practical scenarios is usually carried out by trial and error, since the theoretical developments provide only indicative values. The particularity of the filtered-x structure, which adds a delay between the output of the adaptive filter and the error signal and colors the reference signal, increases the inaccuracy of the theoretical models. Furthermore, the use of band-limited reference signals is beyond the scope of most theoretical studies, which focus on the cases of full-band reference signals, making it difficult to extrapolate these models to band-limited signals. In this paper, we propose a low-cost and meaningful method for selecting the convergence step that is well tuned in most practical cases. It can be used regardless of the type of the reference signal, the system delays, and the nature of the filter required in the filtered-x structure. Numerical simulations are performed to validate the effectiveness of the proposed approach.

СТАБІЛІЗАЦІЯ ДИНАМІЧНИХ ПАРАМЕТРІВ ГРАДІЄНТНОГО ПРОЦЕСУ АДАПТАЦІЇ РАДІОТЕХНІЧНИХ СИСТЕМ З АНТЕННОЮ РЕШІТКОЮ В УМОВАХ АКТИВНИХ ШУМОВИХ ЗАВАД

The article proposes an algorithm for parametric adaptation of radiotechnikal systems (RTS) with an adaptive antenna array (AAR) with stable dynamic parameters (DP) operating under conditions of active noise interference (ANI). When applying gradient methods, not only the convergence of successive approximations is essential, but also their speed, which can vary significantly depending on the existing signal-interference situation. The problem of choosing the convergence step of gradient methods is non-trivial due to the lack of a priori information about the target function which is minimized (maximized). A fixed adaptation step does not guarantee stable convergence and stability of the adaptation process. It is proposed to replace the constant constant of the adaptation step with a variable whose elements depend on the contrast of the spectral distribution of the correlation matrix of observation, and to increase the speed of convergence of these processes, to apply the second-order gradient method. As a result of the synthesis, a generalized analytical form of the quasi-Newton algorithm for an arbitrary step of adaptation for adaptive RTS with an antenna array was obtained. Quasi-Newtonian algorithms for stabilizing the dynamic parameters of the adaptation process of the adaptive interference compensator (AIC) and AAR have been synthesized, which are the best among the algorithms of approximate calculations. The use of these algorithms is advisable in the case of limited computing resources. The conducted study of the efficiency of the quasi-Newtonian algorithm of parametric adaptation for AIC and AAR according to the signal/(interference+noise) ratio criterion, which is achieved at the RTS output, maintains its stability in any of the considered situations, regardless of the influence of arbitrary intensity ANI. Keywords: adaptation, adaptive antenna array, adaptive interference compensator, interference, signal/(interference+noise), quasi-Newton, gradient method.

MIMO Multi-Frequency Active Vibration Control for Aircraft Panel Structure Using Piezoelectric Actuators

Undesirable vibration of aircraft panel structures has a significant impact on the safety of their occupants and the operational reliability of various systems. The traditional time-domain controller based on the parallel multi-frequency filtered-x least mean square (FxLMS) control algorithm is simple and effective. However, the mutual influence of convergence steps and the induced harmonic components will limit its performance. An improved parallel multi-input multi-output multi-frequency FxLMS (MIMOMF-FxLMS) algorithm is proposed, considering the multi-input multi-output (MIMO) control requirement of the aircraft panel structure active vibration control (AVC) system and its inevitable channel coupling phenomenon. The feedback error signal and the control channel model are processed by frequency division in the MIMOMF-FxLMS algorithm, so that the FxLMS algorithm of each control channel runs independently, effectively avoiding mutual interference between convergence steps and the generation of induced harmonic components. Simulation and experimental research on multi-frequency AVC of the panel structure model is performed using a new type of piezoelectric stack actuator. The test results show that the proposed controller reduces the vibration level of each observation point by more than 98%, and the vibration responses of the first two-order frequency component are significantly suppressed, which also has strong adaptability and anti-noise ability.

Protecting-Group-Free Iterative Divergent/Convergent Method for Preparing Sequence-Defined Polymers

A new iterative growth method is reported to prepare monodisperse sequence-defined polymers by combining the chain growth mechanisms of the traditional iterative sequential growth (ISG) and iterative exponential growth (IEG) strategies. Using three orthogonal click reactions to couple three groups of monomers, each containing two orthogonal clickable groups. This method grows monodisperse sequence-defined polymers in a protecting-group-free divergent/convergent way. Each iterative cycle of this method involves two divergent sequential growth steps and one convergent exponential growth step, which endows the new method with the chain growth advantages of both traditional ISG and IEG. The two divergent sequential growth reactions introduce new monomer structures into main chains and provide the resultant sequence-defined polymers with varied monomer sequences, while the convergent exponential growth reaction guarantees the method a fast chain growth speed. The repetition of the abovementioned iterative cycle allows this method to grow monodisperse sequence-defined polymers with a fast chain growth manner of 3 × 2n–2.