The risks associated with the isolated design of complex software systems within individual industries are analyzed, where not only the same thing is often done, but also the quality of the design suffers  due to incomplete competence of the implementers. The approach of dividing competence and responsibility in complex software development by introducing an additional domain-specific layer of interaction between the software developer and the subject area specialists is discussed. The use of an adaptive modelling system as a tool for such separation is proposed. It is shown that the use of adaptive modelling as a common development platform for industry-specific CAD will not only improve the quality of production design in different industries, but will also simplify the design of production in related fields. Finally, it is shown that the use of a common platform will avoid the costs associated with the trend towards simplification and atomization of software developed in our country in the face of sanctions and the degradation of global connections.

The article is devoted to the development of control algorithms for a network of agents of a chain network, each agent of which is a linear plant with state delay, subject to the action of external disturbances under conditions of a priori uncertainty. In each agent of the network, the output of the previous agent is monitored, and the signal from the leading subsystem arrives only at the first agent of the network, the communication is one-way. Taking into account the time delay in the models of each agent of the network of such a structure makes them close to real ones. In agent control systems, disturbances are compensated by implementing the principle of invariance, namely, in each network agent, compensation for the action of external disturbances acting on the network agent from the outside, as well as internal disturbances caused by various modes of operation of the plant, is carried out by generating a special disturbance signal, and then it subsequent compensation with the help of an auxiliary loop and Khalil observers. A numerical example of a chain network consisting of four linear control plants is given under the conditions of interval uncertainty of the parameters of their mathematical models, state delay and the action of external uncontrolled disturbances. Numerical simulation was carried out in Matlab Simulink. Graphs of transient processes for tracking errors of agents of the chain network are presented, confirming the theoretical conclusions and illustrating the good performance of the control algorithms for the chain network.

The paper considers the issues of using comb filters (CF), known in radio engineering, as an alternative to continuous corrective filters, when they are implemented in controllers in automatic control systems. The convenience of their implementation in controllers is associated with the organization of a fixed number of sample passes in the processed signal. In this paper, the properties of two types of comb filters of the first order are highlighted and studied in detail: a high-frequency filter and a proportional (forming) filter. The characteristics of these filters are taken as a basis for the implementation of more complex characteristics of corrective devices in combination with numerical integration algorithms. The analysis of the properties of CF was carried out in the frequency domain. It is shown that the initial sections of the frequency characteristics of the CF at certain filter parameters coincide well with the continuous analogues of such devices. It is these areas of frequency characteristics, limited by the cutoff frequency of the open system, that determine the coincidence of static and dynamic properties of the synthesized system. It is shown that CF reduces on 30 % the level of the standard deviation (RMS) of uniformly distributed white noise at the filter output. The decrease in the level of RMS occurs due to the presence of periodically recurring dips in the amplitude-frequency response (frequency response) of filters forming a so-called comb. Due to the fact that the filters in question "cut out" harmonics in the output signal spectrum that are multiples of the inverse of the total duration of the transmitted reference intervals, then a special choice of this duration can not only reduce the level of the random signal at the CF output, but at the same time significantly reduce the level of regular interference, if it is present. Such filters can be simultaneously used as amplitude modulated signal demodulators. 

The article is devoted to the development of an algorithm for constructing the trajectory of a robotic platform moving in an environment with obstacles. This algorithm is based on the application of a special local optimization procedure at each planning step and allows us to obtain feasible program trajectories without increasing the computational complexity of algorithms compared to existing methods. The algorithm is based on the application of the improved method of potential fields and subsequent smoothing of the resulting trajectory. The improving of the potential field method consists in a new way of detecting and avoiding local minima. When a local minimum is detected, it is added to the map as an additional obstacle, which makes it possible to avoid it during further trajectory planning. To circumvent obstacles that can be approximated by polygons, the method of the effective point to the obstacle is proposed, which is the equivalent of the latter in relation to the current location of the moving robotic platform when using this planning method. A two-stage technique for smoothing piecewise linear trajectories is proposed. It is assumed that there is some initial suboptimal curve found by any planning method. This curve is optimized using a functional that includes the length of the trajectory and the deviation of the optimized curve from the original curve. At the second stage, the linear segments of the planned straight line are conjugated with second-order curves. As a result, the planned trajectory of motion is a quadratic-linear curve with a smooth function of the trajectory velocity. At the same time, the proposed method of coupling rectilinear sections of the trajectory does not require sudden changes in speed when passing turns. Simulation results confirming the effectiveness of the proposed method of planning the trajectories of robots are considered and discussed.

The purpose of the paper is to present the experience of developing algorithms implemented in the AUV control system using artificial intelligence (AI) technologies. The author’s view on the content of the concept of AI in relation to the creation of complex technical systems is presented. Two of the most promising, according to the author, AI technologies are identified: 1) development of the basic algorithm and its improvement based on the results of comprehensive modeling in various operating conditions; 2) creation of a problem-oriented artificial neural network and its deep learning using a large amount of experimentally obtained training material. It is stated that both technologies are quite time-consuming and require a long time to implement. But if the completeness and adequacy of modeling of the conditions in which the system being created is supposed to function plays a key role in modeling technology, then in machine learning technology, the availability of a sufficient amount of training material comes to the fore (in the case of developing vision systems — images of recognizable objects, the number of which can number many thousands).The paper presents the structure of the multi-agent control system of AUV, focuses on the complexity of the tasks it solves and the need to use AI technologies in its creation. It is shown that of all the tasks solved by the AUV control system, AI methods are most in demand for solving two categories of tasks: 1) recognition of the current situation and making an adequate decision in this situation in the interests of completing the route task; 2) search for the designated bottom object among many other bottom objects of natural and artificial origin. The use of AI technologies is demonstrated by the example of the development using a specially created simulation stand of the AUV control algorithm when bypassing an extended. It is proposed to solve the problem of detecting and recognizing a designated bottom object using deep learning technology of a problem-oriented artificial neural network with the peculiarity that the training material is formed programmatically in the form of digital images of the desired bottom object at the output of hydroacoustic, optical and electromagnetic means of monitoring the bottom in various conditions of their observation.

The article is devoted to the study and analysis of the mechanism of undulating propulsion in the case of its application in underwater robots. It gives a brief review of studies related to developing a mathematical model of the movement of an underwater robot. Robot using two-symmetry undulating fin. This mathematical model imitates the movement of some fish species, such as batoid rays. The purpose of the study was to substantiate the possibility of creating a propulsion mechanism. In this mechanism the undulating movement of the fin is provided by individual elements controlled by servomotors. With appropriate control, these movements provide the propagation of the wave in the flexible fin of the robot. The advantage of this approach is a significant increase in the maneuverability of the device compared to existing traditional models. Also, the movement becomes almost silent. The experiments performed on a robot prototype. As a result, a relation is found between the parameters of the undulating movement and the linear speed and thrust of each fin. The parameters of undulating movement such frequency and amplitude are analyzed, to provide stable movement of the underwater robot. The experiments confirmed the possibility of implementing the proposed control method. They Showed the possibilities of controlling the direction of propulsion force by changing oscillating phase of each fin ray. Also, the possibilities of controlling the thrust of the fin by changing the amplitude and frequency of undulating movement of the fin.