AUT JOURNAL OF MODELING AND SIMULATION
Year:2022 | Volume:54 | Issue:1|Papers Count:9

Parallelization is a technique that increases the speed of tasks by processing them simultaneously. Distributed multi-agent systems are one of the cases in which parallel processing techniques can be used. In this paper, first, the multi-agent model of a fuzzy content-based image retrieval system is designed to distribute it. Next, the corresponding parallel multi-agent model is designed. Afterwards, the parallel image retrieval system is implemented on reconfigurable hardware. This method is based on a multi-agent paradigm which is suitable for various parallelisms within applied problems. In this study, by using parallelism techniques, a method is presented that considerably decreases the image retrieval systems consumption time. This paper focuses on how to implement a fuzzy content-based image retrieval system in the form of a multi-agent model. Since reconfigurable hardware is appropriate to support software agents, I also show how these agents use the inherent parallelism of reconfigurable Hardware for parallel image retrieval and increase the speed of this new system greatly. The two sequential and parallel systems are tested on a data set containing 1000 images. The results signify the increase of almost 3-times speed for the proposed parallel system by software agents, and 400-times speed by hardware agents. In order to evaluate the retrieval efficiency of the proposed system compared to the previous works, two other image retrieval systems have been implemented and the efficiency and memory consumption of the systems have been compared. The results indicate better performance of the proposed system than other methods studied.

In recent years, crude oil has been one of the most important energy sources in the world which impacts political stability and economic security in many countries. Furthermore, Crude oil price also has a huge influence on the world economic pattern due to being directly utilized in different industries in various ways. The purpose of this paper is to improve the ability of existing models in forecasting Brent crude oil returns. Hence, we propose two new deep learning-based models. The first model is based on the transformer which has been very popular in Natural Language Processing over the past few years. Moreover, different widely used deep learning-based methods of time series modeling such as SVR, MLP, GPR, and LSTM are implemented. The second model takes the outputs of all implemented methods as new features and feeds them to a Multilayer Perceptron network. The obtained results by each proposed model have been compared together concerning closeness to the real returns according to the predefined metrics. It is demonstrated that the new transformer-based model (the first model) has better results than the other four common machine learning-based methods. Consequently, the new hybrid model (the second model) provides better price forecasts among all implemented models.

Transfer alignment of master and slave systems plays a key role in the inertial navigation accuracy of the marine cooperative vehicles. Accuracy enhancement of misalignment angle and orientation estimation is the main purpose of the transfer alignment. Velocity and orientation matching is a well-known method for transfer alignment. However, in many applications, there are no velocity measurements of both the master and slave systems due to weight, dimensional and technological limitations of accurate speed sensors, such as Doppler Velocity Loggers (DVL). Angular velocity configuration is a suitable solution for transfer alignment in this situation. However, the orientation error cannot be estimated in this configuration. Taking this drawback into account, a new configuration based on using the integral of angular velocity in addition to angular velocity measurement is presented for transfer alignment in the current research. Furthermore, appropriate abilities are considered to estimate the dynamic misalignment angle, orientation error and also measurement errors of the slave gyroscope. Two linear and non-linear observation models are developed for the transfer alignment configuration. The simulation results reveal the appropriate performance of the proposed configuration for marine application, especially when there are no accurate velocity measurements. Based on the simulation results, the performance of the non-linear observation model is better than linear ones in dynamic misalignment angle estimation. Moreover, it can be inferred from the orientation error estimation that rich data in high-maneuvered motion is necessary for required estimation accuracy. Additionally, 200 runs of Monte-Carlo simulation are developed and the estimation RMSE are presented.

In this paper, to solve the output tracking problem of a single-link flexible joint manipulator, Polytopic Linear Models (PLMs) of the dynamics are made to take advantage of this method. Although linear control methods are very useful due to their powerful theories and simplicity, they can only be used in a neighborhood of the equilibrium point. One way to solve this problem is a PLMs-based method that linearizes the dynamics around several operating points. Therefore, in this paper, after calculating the PLMs of the manipulator, a state feedback control is applied to the derived linear dynamics that are augmented with the dynamics of the output tracking error. An extended method is used to decompose the scheduling space to construct PLMs, which is the segregation method improved with an extra aggregation. In order to avoid creating a large number of local models, an axis-oblique decomposition strategy is used instead of an axis-orthogonal decomposition. In addition, the scheduling functions of the PLMs are determined such that overlaps between the regions are avoided. By this selection, the output tracking problem becomes as a Linear Matrix Inequality (LMI) problem instead of a bilinear matrix inequality problem, which is more difficult to solve and may not lead to an optimal global solution.

Facial Expressions are one of the most effective ways for non-verbal communications, which can be expressed as the Micro-Expression (ME) in the high-stake situations. The MEs are involuntary, rapid, subtle, and can reveal real human intentions. However, their feature extraction is very challenging due to their low intensity and very short duration. Although Local Binary Pattern on Three Orthogonal Plane (LBP-TOP) feature extractor is useful for ME analysis, it does not consider essential information. To address this problem, we propose a new feature extractor called Local Binary Pattern from Six Intersection Planes (LBP-SIPl). This method extracts LBP code on Six Intersection Planes, and then combines them. Results show that the proposed method has superior performance in apex frame spotting automatically in comparison with the relevant methods on the CASME I and the CASME II databases. Afterwards, the apex frames are the input of the Fast Region-based Convolutional Neural Network (FR-CNN) to recognize the Facial Expressions. Simulation results show that the ME has been automatically recognized in 81.56% and 96.11% on the CASME I and the CASME II databases by using the proposed method, respectively.

Inertial Navigation Systems suffer from accumulative errors due to their dead reckoning structure. Consequently, the navigation reset or realigning process of such systems is unavoidable. Realigning starts with estimating the navigation errors and correcting the navigation states. To correct the error of attitude states in the navigation reset process, different kinds of attitude correction method are used in the literature. This paper proposed an analytical attitude correction method that can calculate the error of Euler angles more precisely than the conventional method. In addition, this new approach preserves normality and orthogonality characteristics of the transformation matrix while the conventional method leads to losing both of these conditions. The proposed method expresses the error of Euler angles as functions of Euler angles and small rotation angles. The relation between the Euler angles error and the small rotation angles is nonlinear and mathematical calculation is performed to extract the explicit functions. Numerical simulations prove that the proposed method of attitude correction has the mentioned features and its performance is dominant over the conventional method.

Optical Coherence Tomography (OCT) images are used to reveal retinal diseases and abnormalities, such as Diabetic Macular Edema (DME) and Age-related Macular Degeneration (AMD). Fluid regions are the main sign of AMD and DME and automatic fluid segmentation models are very helpful for diagnosis, treatment, and follow-up. This paper presents a two-path Neutrosophic (NS) Fully Convolutional Networks, referred as TPNFCN, as a fully-automated model for fluid segmentation. For this task, OCT images are first transferred to NS domain and then Inner Limiting Membrane (ILM) and Retinal Pigmentation Epithelium (RPE) layers as first and last layers of retina are segmented by graph shortest path algorithms in NS domain, respectively. Afterwards, a basic block of FCN is presented for fluid segmentation and this block is used in the architecture of the proposed TPNFCN. Both the basic block and TPNFCN are evaluated on 600 OCT scans of 24 AMD subjects containing different fluid types. Results reveal that the proposed basic block and TPNFCN outperform five competitive models by improvement of 6.28%, 4.44% and 2.54% with respect to sensitivity, dice coefficients, and precision, respectively. It is also demonstrated that the proposed TPNFCN is robust against low number of training samples in comparison with current models.

The purpose of this paper is to provide a positioning algorithm for high-velocity moving vehicles by a low frequency local positioning system, such as Loren-C navigation system. The performance of the Linear Digital Averaging (LDA) depends on similarity of the reception period of consecutive pulses (i.e. Pulse Code Interval (PCI)). The velocity of the receiver changes the period of pulse reception in each PCI and distorts the average pulse. The distortion of the average pulse depends on the number of pulses and the amount of pulse delays (i.e. the difference between pulse reception period and PCI). In this paper, pulse delay threshold and consequently the velocity of receiver threshold of the acceptable average pulse distortion is analyzed. It is shown that the determined threshold of the velocity of receiver is very low for a wide variety of applications. The proposed solution to increase the velocity threshold is to compensate the pulse delays using the last estimation of the location and the velocity vector of the receiver. The proposed algorithm can be applied to design receivers for high-velocity vehicles. The simulation results confirm the convergence of the proposed positioning algorithm and the feasibility of increasing the velocity threshold by means of pulse delay compensation before the LDA.

The smart city, as the future of urban infrastructure, has a wide range of benefits, including better utilization of scarce resources and more welfare for citizens, and is a necessity for sustainable development. An increasing number of smart city projects have been implemented and many others are under development worldwide. There are many economic, environmental, and social challenges to develop a smart city. One of these challenges is the smartness assessment. Key Performance Indicators (KPIs) are important for measuring and comparing the grade of smart city maturity. The International Telecommunication Union (ITU) has established the United for Smart Sustainable Cities (U4SSC) to standardize and release a KPI set for measuring smart city development from various aspects. In this paper, we expand the standardized KPIs and map it to contextualized KPIs, which concerns the challenges and priorities for the specific case under study. By combining the standardized and contextual KPIs, a comprehensive assessment model is created and used for smart city development. As a case study, we also briefly report the assessment of Mashhad city smartness using the proposed method and the ITU’s U4SSC verification process. We defined the smart tourism and Smart Water Management as the contextualized KPIs for Mashhad and evaluated component of smart tourism and Smart Water Management ecosystem.