Studia Informatica. System and information technology

Artificial Intelligence. Challenges and threats

Andrzej Barczak — 2023-12-26

The paper is concerned with the visions and threats of Artificial Intelligence. It characterizes and describes those that, according to experts and scientists , will have a huge impact on the shape, nature and functioning of societies in the next two to three decades. Undoubtedly, one of the significant challenges is the development and application of deep learning. Scientific research in Artificial Intelligence will be accompanied by intensive development of computer vision, convolutional networks and virtual, augmented and mixed reality. In the future, everything digital will be able to fall prey to counterfeiters. The essence and functioning of related Deepfake tools are explained and described in the paper. The development of Artificial Intelligence will result in a wide and versatile application of Biometrics. The paper describes the prospective areas of its use and draws attention to the problems associated with the construction of the oversized databases necessary for this purpose. In the era of a society characteristic of the development of the fourth industrial revolution, theway and means of automobile transportation and the necessary infrastructure will be radically transformed - autonomous vehicles will appear. The paper describes all levels of reference relating to the scope of control of an autonomous vehicle. Effective use of Artificial Intelligence solutions will not be possible without computers with enormous computing power. Theemergence of quantum computers will certainly solve this problem. The essence of quantum computer functioning is described in the next chapter. Artificial Intelligence can successfully perform many tasks better than humans. This will make Artificial Intelligence take on a huge economic value. Positions will be lost for both physical and mental workers. The question then becomes, where are we? Which professions can take over Artificial Intelligence and which ones will not move? What is the future of work done by humans? These issues are addressed in Chapter 9. The prospect of widespread and comprehensive use of Artificial Intelligence solutions will raise the following question, with respect to both humans and society. Can Artificial Intelligence optimize our happiness? The answer to such a question is not an easy one. The paper attempts to answer it. In the age of Robotics and Artificial Intelligence, we will be accompanied by, revolutionary changes in the way humans of entire societies function. There will be an urgent need to define a neweconomic model. Issues clarifying the nature and importance of the essential components of the said model are addressed in this work. Solutions and use of Artificial Intelligence are often accompanied by various kinds of failures. These can have many causes. The paper describes and explains, proposed by Yampolsky Roman, a kind of dysfunction pattern. The results of scientific research and
practical experience that the list of Artificial Intelligence dangers is very extensive. A representative list of risks from Artificial, along with the characteristics of each of them, is presented in the last part of the paper.

The VR Therapy platform for social anxiety treatment

Waldemar Bartyna — 2023-12-26

During the period of online learning related to the pandemic situation, a cooperation was established between the Institute of Computer Science and the Institute of Pedagogy at Siedlce University of Natural Sciences and Humanities to create a therapeutic environment in virtual reality for the treatment of social anxiety or shyness. Anxiety disorders are classified as the most common mental disorder in children and adolescents, their prevalence is estimated at 2-10 percent of the population. The phenomenon has
increased due to the pandemic situation. The aim of the project is to develop a low-cost and user-friendly technological solution using virtual reality and biofeedback elements for social anxiety therapy, dedicated to children and adolescents aged 12 to 19.

Application of artificial neural networks and NEAT algorithm to control a quadcopter

Mateusz Przychodzki — 2023-12-26

We present a framework for building ArtificialNeuralNetworks (ANNs) able to control a quadcopter and perform basic maneuvers, like hover or following waypoints. In this approach, we make use of the Neuroevolution of Augmenting Topologies (NEAT) algorithm which is aimed at creating the network structure and the weights in result of evolutionary computations. In order to evaluate fitness of individuals, we use physics based, realistic simulation engine Gazebo, where each individual controls a drone in a simulated environment. Our approach is aimed at using one of existing, popular protocols used to remotely control drones, and train ANNs able to imitate signals received from a radio controller operated by a human pilot. Thus, contrary to the most of other approaches, our autonomous controller cooperates with standard drone software. Our ultimate goal is to train ANNs able to control a real-world quadcopter and perform advanced tasks autonomously. Not only such ANNs should be able to perform the maneuvers correctly, but they should be small enough to transfer them into a quadcopter’s limited memory. In this paper we report the first stage of our project - a successful development and deployment of the ANNs distributed training framework, and choosing the activation function for further research.

Towards explainability of hashtags in the light of Graph Spectral Clustering methods

Bartłomiej Starosta — 2023-12-26

Hashtags constitute an indispensable part of modern social media world. As more and more hashtags are invented, it becomes a necessity to create clusters of these hashtags. Nowadays, however, the clustering alone does not help the users. They are asking for justification or expressed in the modern AI language, the clustering has to be explainable. We discuss a novel approach to hashtag explanation via a measure of similarity between hashtags based on the Graph Spectral Analysis. The application of this similarity measure may go far beyond the classical clustering task. It can be used to provide with explanations for the hashtags. In this paper we propose such a novel view of the proposed hashtag similarity measure.

Universal structural map for indoor navigation in university campus

Piotr Świtalski — 2023-12-26

Mapmodeling is an important aspect of indoor navigation. It involves creating a digital map of the indoor environment that can be used for navigation purposes. The map can be created using various techniques such as laser scanning, photogrammetry, and computer vision. Once the map is created, it can be used to develop navigation algorithms that can help users navigate the indoor environment. In this paper, we propose a solution based on CAD files. These models can be used for a variety of purposes, including
indoor navigation. There are many CAD applications available, including AutoCAD, SketchUp, and SolidWorks, among others. It is a relatively cheap method to model any indoor environment from scans of plans or CAD files. CAD files are the most accurate way to build a digital indoor map because these files can include 2D or 3D designs and usually contain important location information (e.g. floor level) within the layer properties. Moreover, we can map the CAD annotation to the following feature classes if they conform to the indoors model: types of rooms, types of doors, etc. We propose a solution based on DXF format files. We developed the parser to transfer the necessary data from the CAD files to the navigation system. It consists of processing existing maps from the CAD format to the appropriate structure, supplementing it with the data of BLE transmitters, and saving it as a graph suitable for determining routes and guiding the user along the route. We use our method in a system that supports navigation and user safety, emphasizing users with special needs, which we are implementing on our academic campus.

An indoor campus navigation system for users with disabilities

Marek Pilski — 2023-12-26

While supporting navigation outside buildings is currentlywell solved through GPS-based systems, only a few solutions support indoor navigation and user positioning in large buildings. In this paper, we propose a solution to support such navigation by creating a dedicated hierarchical map of the building and associating it with the infrastructure of Bluetooth Low Energy (BLE) transmitters installed in the building. Then a user equipped with a smartphone can easily read the signals of these transmitters and, thanks to the map, can determine their position in space.We use our method in a system that supports navigation and user safety, emphasizing users with special needs, which we are implementing on our academic campus. We describe the idea, architecture, and implementation of this system.

Selection of the programming environment for neural modelling of the power and electricity demand generation systems in terms of unmanned factories

Jerzy Tchórzewski — 2023-12-26

The work concerns the selection of the programming language and environment for the needs of neural modeling of the power and electricity demand generation system in terms of uninhabited factories. Therefore, the main goal of the conducted research is to obtain the best possible Artificial Neural Network, i.e. to teach it a model of a real system, which is a system for generating demand for power and electricity based on numerical data on parts of the power system operation in terms of uninhabited factories. The learning capabilities of artificial neural networks were checked by comparing the MSE error and the Regression Index R2. In each of the examined programming languages and related programming environments, i.e. Matlab, Python and Wolfram, an Artificial Neural Network with the same structure and properties was designed and implemented, i.e. with the same number of input and output neurons, the number of hidden layers and the number of neurons in them, the activation function of neurons and the learning method. In addition to the ANN training of the system model, testing and validation as well as comparative studies were carried out.

WPad BVI - a multilingual universal knowledge exchange tool with support for visually impaired users

Dariusz Mikułowski — 2023-12-26

Nowadays, users need to work simultaneously with many files and folders on their computer and various resources on the Internet, including several software pieces. It means that they have many windows open simultaneously, and they must continuously switch between different opened applications such as web browsers, text editors, or audio-visual programs. These problems are also crucial to blind or visually impaired users, for whom this issue is even more difficult. Due to the nature of their work, i.e., they do not use the mouse and cannot see the video and graphic content, they have to remember, at each moment, which programs they have open and which they no longer use. In this paper, we propose an approach that consists of universal software, where, on the one hand, the user can collect and view data; on the other hand, he can produce ready-made output packages in simple web pages. This solution consists of two main parts: an offline softwareWPad and the online - Personal Knowledge Information System (PIKS). We propose a simplified and more automated version of an earlier developed WPad software system called WPad for BVI, which visually impaired persons and other beginners can use. The solution was tested by two BVI users (one BVI student and one BVI teacher) and several non-disabled teachers and students from Poland and Slovakia. Feedback from users was collected that will be used to improve the system further.