Search

The circle packing problem involves finding the best way to place non-overlapping circles within a given space, while the smog-free tower installation problem aims to minimize the exposure of residents to secondhand smog by identifying the optimal tower locations. This study proposes a Real Coded Genetic Algorithm (RCGA) that uses real-valued representations of circle positions to solve the smog-free tower installation problem. A new crossover operator is introduced, combining the information from two parent solutions to generate two new offspring solutions. The operator uses a random crossover point and two scaling factors to control the amount of information exchanged. The performance of the operator is evaluated on CEC-2017 benchmark problem set and compared to other commonly used operators, with results indicating that it produces high-quality solutions and outperforms other operators in terms of solution quality and convergence speed. This research contributes to developing effective optimization algorithms that can have important implications for improving public health and reducing the negative effects of secondhand smog.

The use of unmanned aerial vehicles (UAVs) has been rapidly increasing in both professional and recreational settings, leading to concerns about the safety and security of people and facilities. One area of research that has emerged in response to this concern is the development of detection systems for UAVs. However, many existing systems have limitations, such as detection failures or false detection of other aerial objects, including birds. To address this issue, the development of a standard dataset that provides images of both drones and birds is essential for training accurate and effective detection models. In this context, we present a dataset consisting of images of drones and birds operating in various environments. This dataset will serve as a valuable resource for researchers and developers working on UAV detection and classification systems. The dataset was created using Roboflow software, which enabled us to efficiently edit and manipulate the images using AI-assisted bounding boxes, polygons, and instance segmentation. The software supports a wide range of input and output formats, making it easy to import and export the dataset in different machine learning frameworks. To ensure the highest possible accuracy, we manually segmented each image from edge to edge, providing the YOLO model with detailed and accurate information for training. The dataset includes both training and testing sets, allowing for the evaluation of model performance and accuracy. Our dataset offers several advantages over existing datasets, including the inclusion of both drones and birds, which are commonly misclassified by detection systems. Additionally, the images in our dataset were collected in diverse environments, providing a wide range of scenarios for model training and testing. The presented dataset provides a valuable resource for researchers and developers working on UAV detection and classification systems. The inclusion of both drones and birds, as well as the diverse range of environments and scenarios, makes this dataset a unique and essential tool for training accurate and effective models. We hope that this dataset will contribute to the advancement of UAV detection and classification systems, improving safety and security in both professional and recreational settings.

Abstract Multiple attention-based models that recognize objects via a sequence of glimpses have reported results on handwritten numeral recognition. However, no attention-tracking data for handwritten numeral or alphabet recognition is available. Availability of such data would allow attention-based models to be evaluated in comparison to human performance. We collect mouse-click attention tracking data from 382 participants trying to recognize handwritten numerals and alphabets (upper and lowercase) from images via sequential sampling. Images from benchmark datasets are presented as stimuli. The collected dataset, called AttentionMNIST, consists of a sequence of sample (mouse click) locations, predicted class label(s) at each sampling, and the duration of each sampling. On average, our participants observe only 12.8% of an image for recognition. We propose a baseline model to predict the location and the class(es) a participant will select at the next sampling. When exposed to the same stimuli and experimental conditions as our participants, a highly-cited attention-based reinforcement model falls short of human efficiency.

The paper introduced a novel approach for automatic assessment of olfactory perceptual-ability of human-subjects using a functional Near Infrared Spectroscopy device. The assessment requires fuzzy functional mapping from spectroscopic measurement to perceptual-ability using Type-2 fuzzy reasoning. The novelty of the work lies in Vertical Slice Based General Type-2 Fuzzy Reasoning which employs fuzzy meet and union between the planes of type-2 measurement and observation spaces using the classical definition of t-norms and s-norms. The results of the meet and the union computation are later used as the Lower and Upper Firing Strength of the fired rule to determine the structure of the inference. Experiments undertaken confirm the efficacy of the proposed technique over traditional functional mapping, involving neural networks, regression analysis, and the like. The proposed technique of olfactory perceptual-ability can be directly employed to determine the thresholds for recognition-probability and discrimination-probability, when submitted to the subject in presence of aromatic noise. An analysis is undertaken to measure the computational overhead, which is found of the order of $O(m.n)$ and run-time complexity of 94.78 ms, where $m$ and $n$ respectively represent discretizations in the vertical slice and features respectively. A statistical test undertaken confirms the superior performance of the proposed system with others at 95% confidence level.

This theoretical study analyses the effects of geometrical and fluid parameters on the flow metrics in the Hagen-Poiseuille and plane-Poiseuille flows of Herschel-Bulkley fluid through porous medium which is considered as (i) single pipe/single channel and (ii) multi–pipes/multi-channels when the distribution of pores size in the flow medium are represented by each one of the four probability density functions: (i) Uniform distribution, (ii) Linear distribution of Type-I, (iii) Linear distribution of Type-II and (iv) Quadratic distribution. It is found that in Hagen-Poiseuille and plane-Poiseuille flows, Buckingham-Reiner function increases linearly when the pressure gradient increases in the range 1 - 2.5 and then it ascends slowly with the raise of pressure gradient in the range 2.5 - 5.In all of the four kinds of pores size distribution, the fluid’s mean velocity, flow medium’s porosity and permeability are substantially higher in Hagen-Poiseuille fluid rheology than in plane-Poiseuille fluid rheology and, these flow quantitiesascend considerably with the raise of pipe radius/channel width and a reverse characteristic is noted for these rheological measures when the power law index parameter increases.The flow medium’s porosity decreases rapidly when the period of the pipes/channels distribution rises from 1 to 2 and it drops very slowly when the period of the pipes/channels rises from 2 to 11.

Real-world applications of intelligent agents demand accuracy and efficiency, and seldom provide reinforcement signals. Currently, most agent models are reinforcement-based and concentrate exclusively on accuracy. We propose a general-purpose agent model consisting of proprioceptive and perceptual pathways. The agent actively samples its environment via a sequence of glimpses. It completes the partial propriocept and percept sequences observed till each sampling instant, and learns where and what to sample by minimizing prediction error, without reinforcement or supervision (class labels). The model is evaluated by exposing it to two kinds of stimuli: images of fully-formed handwritten numerals and alphabets, and videos of gradual formation of numerals. It yields state-of-the-art prediction accuracy upon sampling only 22.6% of the scene on average. The model saccades when exposed to images and tracks when exposed to videos. This is the first known attention-based agent to generate realistic handwriting with state-of-the-art accuracy and efficiency by interacting with and learning end-to-end from static and dynamic environments.

Abstract This work analyses the performance of Reinforcement Learning (RL) versus Swarm Intelligence (SI) for coordinating multiple unmanned High Altitude Platform Stations (HAPS) for communications area coverage. It builds upon previous work which looked at various elements of both algorithms. The main aim of this paper is to address the continuous state-space challenge within this work by using partitioning to manage the high dimensionality problem. This enabled comparing the performance of the classical cases of both RL and SI establishing a baseline for future comparisons of improved versions. From previous work, SI was observed to perform better across various key performance indicators. However, after tuning parameters and empirically choosing suitable partitioning ratio for the RL state space, it was observed that the SI algorithm still maintained superior coordination capability by achieving higher mean overall user coverage (about 20% better than the RL algorithm), in addition to faster convergence rates. Though the RL technique showed better average peak user coverage, the unpredictable coverage dip was a key weakness, making SI a more suitable algorithm within the context of this work.

Graph grammars can be used to model the development of diverse graph families. Since their creation in the late 1960s, graph grammars have found usage in a variety of fields, such as the design of sophisticated computer systems and electronic circuits, as well as visual languages, computer animation, and even the modelling of intricate molecular structures Replacement of edges and nodes are the two primary approaches of graph rewriting. In this paper we introduce a new type of node replacement graph grammar known as nc-eNCE graph grammar. With this new class of graph grammars we generated certain graph classes and we showed that these class of graph grammars are more powerful than the existing edge and node controlled embedding graph grammars. In addition, these graph grammars were used to model several common protein secondary structures such as parallel and anti-parallel β-sheet structures in different configurations. The use of these graph grammars in modelling other bio-chemical structures and their interactions remains to be explored.

Relating graph structures with words which are finite sequences of symbols, Parikh word representable graphs (PWRGs) were introduced. On the other hand, in chemical graph theory, graphs have been associated with molecular structures. Also, several topological indices have been defined in terms of graph parameters and studied for different classes of graphs. In this study, we derive expressions for computing certain topological indices of PWRGs of binary core words, thereby enriching the study of PWRGs.

A word is a finite sequence of symbols. Parikh matrix of a word is an upper triangular matrix with ones in the main diagonal and nonnegative integers above the main diagonal which are counts of certain scattered subwords in the word. On the other hand, a picture array, which is a rectangular arrangement of symbols, is an extension of the notion of a word to two dimensions. Parikh matrices associated with a picture array have been introduced, and their properties have been studied. Here, we obtain certain algebraic properties of Parikh matrices of binary picture arrays based on the notions of power, fairness, and a restricted shuffle operator extending the corresponding notions studied in the case of words. We also obtain properties of Parikh matrices of arrays formed by certain geometric operations.

Recently, swarm intelligence-based algorithms gained attention of the researchers due to their wide applicability and ease of implementation. However, much research has been made on the development of swarm intelligence algorithms but theoretical analysis of these algorithms is still a less explored area of the research. Theoretical analyses of trajectory and convergence of potential solutions toward the equilibrium point in the search space can help the researchers to understand the iteration-wise behavior of the algorithms which can further help in making them efficient. Artificial bee colony (ABC) optimization algorithm is swarm intelligence-based algorithm. This paper presents the convergence analysis of ABC algorithm using theory of dynamical system. Convergent boundaries for the parameters of ABC update equation have also been proposed. Also, the trajectory of potential solutions in the search space is analyzed by obtaining a partial differential equation corresponding to the position update equation of ABC algorithm. The analysis reveals that the ABC algorithm performs better when parameters of the update equation are in the convergent region and potential solutions movement follows 1-D advection equation.

In the bio-inspired area of membrane computing, a novel computing model with a generic name of P system was introduced around the year 2000. Among its several variants, string or array language generating P systems involving rewriting rules have been considered. A new picture array model of array generating P system with a restricted type of picture insertion rules and picture array objects in its regions, is introduced here. The generative power of such a system is investigated by comparing with the generative power of certain related picture array grammar models introduced and studied in two-dimensional picture language theory. It is shown that this new model of array P system can generate picture array languages which cannot be generated by many other array grammar models. The theoretical model developed is for handling the application problem of generation of patterns encoded as picture arrays over a finite set of symbols. As an application, certain floor-design patterns are generated using such an array P system.

Tile Pasting P System is a computational model, based on the P System model, to generate two-dimensional tiling patterns using pasting rules at the edges of the regular polygons. Computational mechanism plays an important role in understanding the various complexities involved in the formation of complex patterns. In this paper we study the construction of non-periodic tiling patterns using corner tiles and aperiodic Wang tiles using the computational model Tile Pasting P System. We show that the Tile Pasting P System requires a minimum of four membranes to generate the non-periodic Wang tiling.

The problem of finding the global minimum of molecular potential energy function is very challenging for algorithms which attempt to determine global optimal solution. The principal difficulty in minimizing the molecular potential energy function is that the number of local minima increases exponentially with the size of the molecule. The global minimum of the potential energy of a molecule corresponds to its most stable conformation, which dictates the majority of its properties. In this paper the efficiency of four newly developed real coded genetic algorithms is tested on the molecular potential energy function and their supremacy is established over other existing algorithms. The minimization of the function is performed on an independent set of internal coordinates involving only torsion angles. Computational results with up to 100 degrees of freedom are presented.

In this paper, a low cost, wearable six Degree of Freedom (6-DOF) hand pose tracking system is proposed for Virtual Reality applications. It is designed for use with an integrated hand exoskeleton system for kinesthetic haptic feedback. The tracking system consists of an Infrared (IR) based optical tracker with low cost mono-camera and inertial and magnetic measurement unit. Image processing is done on LabVIEW software to extract the 3-DOF position from two IR targets and Magdwick filter has been implemented on Mbed LPC1768 board to obtain orientation data. Six DOF hand tracking outputs filtered and synchronized on LabVIEW software are then sent to the Unity Virtual environment via User Datagram Protocol (UDP) stream. Experimental results show that this low cost and compact system has a comparable performance of minimal Jitter with position and orientation Root Mean Square Error (RMSE) of less than 0.2 mm and 0.15 degrees, respectively. Total Latency of the system is also less than 40 ms.

The steady flow of Herschel-Bulkley and Casson fluids for blood flow in tubes filled with homogeneous porous medium with (i) constant and (ii) variable permeability is analyzed. The expression for the shear stress is obtained first by general iteration method and then using numerical integration; the solutions for velocity and flow rate are obtained. It is noticed that the shear stress and plug core radius are considerably higher in the case of variable permeability than those of the constant permeability case. The velocity and flow rate of both the fluids increase considerably with the increase in the permeability factor and they decrease with the increase in the yield stress of the fluids. The velocity and flow rate of Herschel-Bulkley fluid are considerably higher than those of Casson fluid. Aforesaid flow quantities are significantly higher for flow in tubes with variable permeability than for flow in tubes with constant permeability.

Pulsatile flow of blood in constricted narrow arteries under periodic body acceleration is analyzed, modeling blood as non-Newtonian fluid models with yield stress such as (i) Herschel-Bulkley fluid model and (ii) Casson fluid model. The expressions for various flow quantities obtained by Sankar and Ismail (2010) for Herschel-Bulkley fluid model and Nagarani and Sarojamma (2008), in an improved form, for Casson fluid model are used to compute the data for comparing these fluid models. It is found that the plug core radius and wall shear stress are lower for H-B fluid model than those of the Casson fluid model. It is also noted that the plug flow velocity and flow rate are considerably higher for H-B fluid than those of the Casson fluid model. The estimates of the mean velocity and mean flow rate are considerably higher for H-B fluid model than those of the Casson fluid model.

A new treatment for homotopy perturbation method is introduced. The new treatment is called He-Laplace method which is the coupling of the Laplace transform and the homotopy perturbation method using He’s polynomials. The nonlinear terms can be easily handled by the use of He’s polynomials. The method is implemented on linear and nonlinear partial differential equations. It is found that the proposed scheme provides the solution without any discretization or restrictive assumptions and avoids the round-off errors.