The paper presents the analysis of modern Artificial Intelligence algorithms for the automated system supporting human beings during their conversation in Polish language. Their task is to perform Automatic Speech Recognition (ASR) and process it further, for instance fill the computer-based form or perform the Natural Language Processing (NLP) to assign the conversation to one of predefined categories. The State-of-the-Art review is required to select the optimal set of tools to process speech in the difficult conditions, which degrade accuracy of ASR. The paper presents the top-level architecture of the system applicable for the task. Characteristics of Polish language are discussed. Next, existing ASR solutions and architectures with the End-To-End (E2E) deep neural network (DNN) based ASR models are presented in detail. Differences between Recurrent Neural Networks (RNN), Convolutional Neural Networks (CNN) and Transformers in the context of ASR technology are also discussed.

Odor source location technology has important application value in environmental monitoring, safety emergency and search and rescue operations. For example, it can be used in post-disaster search and rescue, detection of hazardous gas leakage, and fire source detection. Existing odor source location methods have problems such as low search efficiency, inability to adapt to complex environments, and inaccurate odor source location. In this study, based on unmanned aerial vehicle technology and using swarm intelligence optimization algorithm, an improved artificial fish swarm algorithm (IAFSA) is proposed by combining curiosity in psychology on the basis of retaining the good optimization performance of the artificial fish swarm algorithm. The algorithm quantifies the curiosity of artificial fish searching high-concentration areas through a model, dynamically adjusts the artificial fish field of vision and step length with the calculated curiosity factor, and avoids the oscillation phenomenon in the later stage of the algorithm. Simulation results show that the IAFSA has a higher success rate and smaller location error. Finally, odor source location experiments were carried out in an indoor physical environment, the feasibility of the odor source location method proposed in this study is verified in actual scenarios.

This paper presents a concept of architecture and ontology layouts for the development of multiagent model-based predictive control systems. The presented architecture provides guidelines to simplify the development of agent-based systems and improve their maintainability. The proposed multiagent system (MAS) layout is split into multiple subsystems that include agents dedicated to performing assigned tasks. MAS implementation was prepared which can use provided algorithms and actuators and can react to changes in its environment to reach the best available control quality. An example of MAS based on the proposed architecture is shown in the application of dissolved oxygen (DO) concentration control in a laboratory-activated sludge setup with a biological reactor. For that application, MAS incorporates agent-based controllers from the boundary-based predictive controllers (BBPC) family. Presented experiments prove the flexibility, resilience, and online reconfiguration ability of the proposed multiagent system.

Cloud computing has become ubiquitous in modern society, facilitating various applications ranging from essential services to online entertainment. To ensure that quality of service (QoS) standards are met, cloud frameworks must be capable of adapting to the changing demands of users, reflecting the societal trend of collaboration and dependence on automated processing systems. This research introduces an innovative approach for link prediction and user cloud recommendation, leveraging nano-grid applications and deep learning techniques within a cloud computing framework. Heuristic graph convolutional networks predict data transmission links in cloud networks. The trust-based hybrid decision matrix algorithm is then employed to schedule links based on user recommendations. The proposed model and several baselines are evaluated using real-world networks and synthetic data sets. The experimental analysis includes QoS, mean average precision, root mean square error, precision, normalized square error, and sensitivity metrics. The proposed technique achieves QoS of 73%, mean average precision of 59%, root mean square error of 73%, precision of 76%, normalized square error of 86%, and sensitivity of 93%. The findings suggest that integrating nano-grid and deep learning techniques can effectively enhance the QoS of cloud computing frameworks.

The paper presents a metamaterial with reconfigurable effective properties, intended to operate in alternating magnetic fields. The structure of a resonator, based on a series connection of a planar inductor and a lumped capacitor, is expanded using an additional capacitor with a MOSFET transistor. Due to the presence of the controllable active element, it is possible to dynamically change the phase of the current flowing through a meta-cell and shift a frequency response within an assumed range. Since the transistor is driven by the unipolar square wave with a changeable duty cycle and time delay, two closed-loop controllers were utilized to achieve a smart material, able to automatically attain and maintain the imposed resonant frequency. As a result, the complex effective magnetic permeability of the metamaterial can be smoothly changed, during its operation, via an electrical signal, i.e. by adjusting the parameters of a control signal of the active element. The design of the meta-cell, as well as the measuring, data operation and control part, are presented in detail. An illustrative system is examined in terms of achieving the user-defined resonance point of the metamaterial. Transient responses with estimated settling times and steady-state errors and the effective permeability characteristics for the exemplary cases are shown. The meta-cell is also tested experimentally to validate the theoretically determined effective properties.