Natural and man-made disasters can severely destroy environments and they make conditions difficult to access the affected areas and to provide assistance. The conditions on-site could be dangerous and unstable and there is an increasing need for life-saving decisions to be taken quickly to minimize evolving hazards and to start relief operations. The accurate and timely data gathering is important to produce a full information about the calamity. In recent disasters drones are deployed extensively to help find people quickly, provide imminent reliable imagery and data by flying closer to the ground. They are used to create disaster maps and assess damage after earthquakes, landslides, hurricanes, etc. The fast transition into the digital age makes new technologies become available to enhance and expand drone capabilities in disaster risk management, such as Virtual Reality (VR) and Augmented Reality (AR). The paper tries to analyze how VR can be used to plan operations in a controlled manner before deadly events strike by creating disaster simulations in digital environments, enabling the rescuers to practice as many times as necessary until they are able to achieve mastery of the life-saving techniques. The paper also analyzes how drones, equipped with cameras, devices and AR, can be used to create different types of maps that help rescuers locate critical spots. These can also facilitate the location of people in need, and can survey constructions to find critical damages.

The paper discusses the approach to the training of ergatic system operators, which implies the application of virtual reality elements. It presents the review of the existing models of the virtual environment application, offers a system model of the operator's response, systemizes the main structural and functional elements of personal experience in virtual environments.

The paper presents an overview of contemporary approaches to application of virtual reality systems for training of operating personnel for man-machine systems. Special emphases are made on application issues of virtual and alternate reality systems connected with probable cognitive dissonance in operation of real physical objects. The prospects of virtual reality application in professional training are outlined.

A two-stage GridSearch combined with active learning was employed to optimize GAP model parameters for Mg clusters, enabling reliable structural predictions in the extrapolative domain Mg_n, n > 50.