In this paper we consider two families of competing algorithms for finding the shortest paths between all pairs of vertices (APSP) in directed weighted large graphs with different edge densities: Dijkstra and Floyd-Warshall. For comparison, we have taken Dijkstra's algorithm with dynamically varying binary heap, which solves the APSP problem purely in parallel by repeatedly executing on all vertices of the graph considered as source vertices, and we have taken blocked Floyd-Warshall algorithm, which is also well-parallelizable. It is known that in terms of computational complexity, the first algorithm is preferable on sparse graphs and the second algorithm is preferable on dense graphs. At the same time, it is not clear what are the ranges of graph densities at which the first algorithm will consume less CPU time than the second algorithm. This paper describes multithreaded implementations of parallel algorithms on multicore processors that make different usage of synchronization primitives such as mutex, conditional variable, locking, and atomic operation. By conducting computational experiments on an 8-core Intel(R) Core(TM) i7-10700 CPU @ 2.90GHz, we found that each algorithm has a preferred graph density. In the case of multithreaded parallel implementation, the blocked Floyd-Warshall algorithm has lower running time than Dijkstra's algorithm if the graph density is greater than 0.5. Otherwise, Dijkstra's algorithm runs faster. In the case of single-threaded implementation, the split point is 0.43.

Training a relatively big neural network within the framework of deep reinforcement learning that has enough capacity for complex tasks is challenging. In real life the process of task solving requires system of knowledge, where more complex skills are built upon previously learned ones. The same way biological evolution builds new forms of life based on a previously achieved level of complexity. Inspired by that, this work proposes ways of increasing complexity, especially a way of training neural networks with smaller receptive fields and using their weights as prior knowledge for more complex successors through gradual involvement of some parts, and a way where a smaller network works as a source of reward for a more complicated one. That allows better performance in a particular case of deep Q-learning in comparison with a situation when the model tries to use a complex receptive field from scratch.

The method used to calculate natural frequencies and vibration shapes is described. Code fragments for construction for some types of filler are presented. The obtained frequencies and shapes of natural vibrations for three types of filler on the example of one construction are given.

A method has been proposed, and a computer program has been developed for simulating obstacle avoidance by a robotic system. To expand the mobility capabilities of the robotic system, the algorithm combines the core elements of the vfh algorithm and our previously developed algorithm I [1], and is positioned in this work as algorithm II. An evaluation of the average obstacle avoidance time for various types of obstacles has been conducted, including five cubic obstacles, a long wall obstacle, and a complex obstacle. During the computational experiment using the Gazebo 11 simulation environment, the time parameters for robot movements around various types of obstacles were calculated, including five cubic obstacles, a wall obstacle, and a complex obstacle. In the computational experiment, statistical processing of the obtained results was carried out. It was shown that using the proposed algorithm, the assessment of the average time for the robotic system to avoid obstacles is reduced in some cases by up to 7.2 times compared to the use of algorithm I.

The moistening of raw cotton and cotton fiber is a critical stage in the textile production process, significantly influencing the quality of the final product. However, achieving optimal moisture content presents numerous challenges. One of the primary problems is the uneven distribution of moisture during processing, which can lead to variations in fiber strength and quality. The inappropriate application of water or humidification methods often causes fiber shrinkage, reduced durability, and increased contamination. Furthermore, the control of temperature and humidity plays a vital role in preventing over-drying or excessive moisture retention, both of which can negatively affect the spinning and handling properties of cotton fibers. Technological limitations, coupled with the complexity of monitoring moisture levels accurately, result in inefficiencies in the cotton processing industry. Research has also shown that improper moisture regulation can lead to decreased efficiency in machinery, increased energy consumption, and a higher likelihood of fiber damage during transportation and storage. Therefore, addressing these issues requires advancements in moisture control technologies and the development of standardized procedures to ensure consistent and high-quality cotton fiber production.

Modern solar power plants are usually built in deserts, in conditions of lack of water and a lot of dust. Dustiness of solar panels leads to a decrease in their efficiency, and cleaning the panels from dust is an energy consuming operation. The article discusses an approach to assessing the degree of contamination of a panel and making a decision about the need to clean it based on analysis of the panel image. An algorithm for making a cleaning decision is presented, a software application for assessing the degree of contamination of the panel is described, and the results of an experimental test of the created application are presented.

The article presents a comparative analysis of two variants of symmetrical placement of omnidirectional mecanum wheels in a mobile robot. The modeling of the mobile robot behavior with two symmetrical variants of mecanum wheel arrangement along curvilinear trajectories in the MATLAB Simulink environment is considered. Graphs of the robot center trajectories in the X–Y axes are obtained for two variants of symmetrical mecanum wheel configurations when moving along the trajectories of a lemniscate, ellipse, hypotrochoid, and track. Dependences of the robot position errors in the X–Y axes on time are obtained for two variants of wheel topologies when moving along four trajectories. Dependences of the robot body rotation angle deviation on time are analyzed for two variants of wheel topologies when moving along four curvilinear trajectories. The comparative analysis made it possible to give recommendations on movement along the ellipse and lemniscate, which are characterized by smaller position errors and deviations in the rotation angle compared to more complex curves.

The purpose of this paper is to develop a simple criterion for image quality assessment of a signature scanned from a paper (i. e., a static or offline signature). A new approach to no-reference assessment of the quality of a binary signature image is proposed, which can be useful as a tool for monitoring signature samples in biometric recognition systems to control data quality. For example, it can be used during image registration, selection of processing methods and its parameters adjustment, after performing various operations (such as rotation or scaling) and the need to evaluate and analyze the obtained signature images. The paper also describes factors that can negatively affect the quality of a static signature. The experimental analysis was performed on digital images of signatures available in the CEDAR, BHSig260-Bengali, SigComp2009 databases and images collected during the research.