The following paper describes a prototype design for a moving platform to capture images of the spine during walking. The platform is intended to provide a noninvasive method for diagnosing spinal disorders and evaluating the effectiveness of therapeutic interventions. The prototype consists of a treadmill-like structure with a camera mounted above the walking surface, which captures images of the spine from various angles. The platform is designed to move synchronously with the subjects gait, thereby minimizing motion artifacts and producing high-quality images. This moving platform has the potential to provide valuable insights into the dynamics of the spine during walking. In addition, it could be used to investigate the effects of different walking patterns or interventions on spinal biomechanics. The prototype is designed to be portable and can be easily transported to different locations for use in research and clinical settings.
The problem of radar radiation pattern formation using a sparse linear antenna array is considered. An overview of methods for forming virtual structures is presented. The shape and properties of the side lobes of amplitude radiation patterns for co-prime structures are studied using concrete examples. A comparison of the amplitude radiation patterns of the obtained sparse grating and headlights is carried out.
Based on the needs of the practice of ensuring the acoustic safety of the professional activities of the engineering and technical staff of aviation, the specific features of aircraft noise, the experience in the development and operation of personal protective equipment against aircraft noise, the image of the information-measuring system for qualimetry of personal protective equipment against aircraft noise has been developed.
The paper deals with multi-drive industrial mechanisms (units) with non-linear elements (friction, dead zones, etc.). Control systems of electric drives are affected by random perturbing influences. In order to ensure the required dynamic characteristics of control systems containing several actuators, controllers that compensate these perturbations are synthesized. In order to solve this problem, a methodology has been developed to synthesize controllers based on a neural network (NN) with learning.
The problem of distributed storage of big data on clusters is discussed. Known methods of distributed storage are considered: fragmentation and replication. Replication schemes are considered that allow accelerating access to data: peer-to-peer and master-slave. It is shown that replication leads to data inconsistency, which in practice is a compromise formulated in the CAP theorem: out of three properties-data consistency and availability, their partition resistance -no more than two properties can be provided. A model for determining the required number of nodes in a cluster and the number of replicas based on the queuing theory is proposed. In terms of solving the fragmentation problem, a recurrent procedure and a method for choosing the optimal cluster structure are proposed. As part of solving the replication problem, a model for choosing the number of replicas in the event of a “write-write” conflict in the “master-slave” scheme is proposed.
