Farmers explore the capabilities for applications of Robotic Process Automation (RPA) with image processing, pattern recognition and machine learning, so its logical to ask where best to apply this technology for maximum effect. Machines can do automated tasks better, cheaper and faster. One can use the camera from the sky and measure things on the earth especially on cropland from drones and satellites. Over time, drones have increased in capabilities and fallen in cost, and their use has greatly expanded especially in complex terrain. High quality remote sensing with spectral imaging using drones makes them interesting for regular use in Precision Agriculture (PA). Drones are often used in agriculture in ways that were highly controversial only a short time ago even there are no unified legislation on drones usage in agriculture. In this paper we address problem of Remote Sensing (RS) technologies combined with Unmanned Aerial System (UAS) platforms to support and develop selected agriculture operations like map or sensor-based Variable Rate Application (VRA).

Out of the many computing and software oriented models that are being adopted by Computer Networking, Fog Computing has captured quite a wide audience in Research and Industry. There is a lot of confusion on its precise definition, position, role and application. The Internet of Things (IOT), todays' digitized intelligent connectivity domain, demands real time response in many applications and services. This renders Fog Computing a suitable platform for achieving goals of autonomy and efficiency. This paper is a justification of the concepts, interest, approaches, and practices of Fog Computing. It describes the need for adopting this new model and investigate its prime features by elucidating the scenarios for implementing it, thereby outlining its significance in the IoT world.

In recent years there has been a growing interest in wireless sensor networks (WSN) applications. Such sensor networks can be used to control temperature, humidity, contamination, pollution etc. Self-organization and routing algorithms dedicated to wireless sensor networks usually assume that sensors absolute positions are unknown and all decisions are based on sensor's own local information. This assumption makes wireless sensor networks more flexible and energy conserve because making decisions locally is faster and energy efficient. But sooner or later sensors positions have to be found (when sensor sends a message about some event we of course would like to know where this event takes place). In this paper we describe different solutions of finding transceivers positions in wireless networks and we discuss localization in wireless sensor networks. We propose to transfer localization function from base stations to every sensor. We evaluate presented method using simulations

The paper presents a comprehensive overview of intelligent video analytics and human action recognition methods. The article provides an overview of the current state of knowledge in the field of human activity recognition, including various techniques such as pose-based, tracking-based, spatio-temporal, and deep learning-based approaches, including visual transformers. We also discuss the challenges and limitations of these techniques and the potential of modern edge AI architectures to enable real-time human action recognition in resource-constrained environments.

Accurately solving discrete optimization problems is still a serious challenge for classical computing systems. Despite significant progress, it is still impossible to optimally solve examples of the size found in real-world applications. This article examines the problem of scheduling tasks on one machine. Each task has an associated execution time, desired completion date, and a late penalty feature weight. The order of executing tasks should be determined in a manner that minimizes the sum of delay costs. Despite the simplicity of the formulation, this problem is NP-hard and is representative of optimization challenges encountered in the practice of production control, management, and logistics.In this work, we propose using a quantum computing environment to solve it. The main disadvantage of calculations on quantum computers is their nondeterminism. Therefore, there is no guarantee of optimality of solutions to discrete optimization problems determined on a quantum computer.We propose a novel approach to constructing hybrid algorithms that guarantee the optimality of the solution. The algorithm for solving the task scheduling problem considered here is based on the branch and bound method. Calculations are performed alternately on a classical computer and a quantum computer. The lower and upper bounds of the criterion function are determined on a quantum computer. Computational experiments were performed on the D-Wave quantum computer. In small examples, the algorithm runs faster and generates significantly fewer solution tree vertices than its counterpart running on a classic computer. For larger examples, the proposed runtime-constrained method, acting as a heuristic, is an interesting alternative to using quantum computing hardware in relation to classical approaches based on silicon computers.