No. 1 (2020)

The problem of calculation of an autonomous robotic group in order to destroy the detected enemy group is considered. A group of robots must be formed in such a way that the task assigned to it to destroy the enemy group is performed with a high degree of probability. The task is solved as an assignment problem. The initial information for solving this problem are types and numberof objects of the detected enemy group, positions of the enemy objects, information about the war possibilities of the tools available in our group, the type of group being formed (robotic or mixed), the purpose of the operation, the actions of the group at the end of the operation. We propose a solution to the problem based on the evaluation of the effectiveness of individual robotic systems. The solution is formulated as a sequence of the four stages. At the first stage, the calculation of a priori effectiveness of each element of the detected enemy group is performed. At the second stage, based on expert assessments, the choice of efficiency coefficients for each of the available robotic systems against each element of the detected enemy group is made. At the third stage, a priori estimates of the effectiveness of the available robotic systems are corrected, taking into account the coefficients selected at the second stage. At the fourth stage, a group of robotic systems is formed in such a way that its total application efficiency exceeds the total application efficiency of the detected enemy by 2.0–2.5 times. The proposed method of forming a group allows you to cre-ate both quantitative and qualitative composition of the group. The article provides an example of the formation of a group whose goal is to neutralize an exposed enemy.

Unmanned helicopters perform many tasks in difficult operating conditions and are subject to various destabilizing factors that significantly affect flight safety. The main problems encoun-tered in the operation of unmanned helicopters are considered. It is shown that the insufficient level of flight safety is caused, in particular, by the high frequency of crashes during forced land-ings. The necessity of creating onboard means of automatic landing of an unmanned helicopter is proved. Taking into account the requirements of the Federal aviation regulations for landing plac-es, the parameters-restrictions that allow formalizing the choice of terrain areas suitable for land-ing according to the onboard technical vision system are formulated. On the basis of comparative analysis, it is shown that at present, when forming the initial video data for solving this problem, it is advisable to use a complex system of technical vision based on mutually adjusted and having a common viewing area of a 3D laser sensor, color video camera and thermal imager. The proposedrecognition algorithms of the pick-up location in the video data on-Board complex system of tech-nical vision with the use of geometric criteria and the reference permeability. It is proposed to perform the recognition of landing places based on the criterion of geometric cross-country capa-bility in two stages: at the first stage, a map of terrain heights is formed based on 3D laser sensor data, and at the second stage, areas suitable for helicopter landing are selected. Recognition of suitable and unsuitable areas is performed by comparing the elevation differences of this terrain with the reference elevation differences defined for this unmanned helicopter. It is proposed to perform the recognition of suitable landing sites based on the criterion of reference passability by calculating the Euclidean distance between the obtained data and pre-known standards corre-sponding to different types of soil in the six-dimensional feature space (height variance, reflected signal intensity, three colors, and temperature). The final selection of suitable places for planting is proposed to be made from sites that meet both criteria. The results of the work of the corre-sponding software and hardware in real conditions are presented, confirming the correctness and effectiveness of the proposed algorithms.

The paper presents the tasks to be solved by promising RTC VN in the interests of Rvi. The conclusion is formulated that the problem of training and improving the quality of knowledge of Rvi specialists using military robotic systems remains one of the urgent problems of higher mili-tary professional education and acquires new aspects of consideration. It is shown that one of the effective ways to solve the problem of training and improving the quality of knowledge of Rvi spe-cialists is to develop and implement computer-based training systems with elements of virtual real-ity and 3D visualization of the studied samples of equipment and weapons in the educational pro-cess. The main features of the computer information and reference system "compendium of the Rvi" developed at the Mikhailovsky military artillery Academy and used in the educational process are briefly described. Preliminary results of the conducted pedagogical experiment with the use of the "Rvi compendium" are presented, and the main factors that increase the effectiveness of the educational process are Noted. Based on the results of the pedagogical experiment made a rea-sonable inference that the use of KISS "compendium Rvia" allows to increase learning efficiency, to reduce terms of development of technology that is more efficient use of training time and as a result reduce the cost of training and the number of vehicles.

The problem of controlling the interaction of unmanned aerial vehicles (UAVs) with ground-based robotic service platforms that perform the functions of transporting and transferring physi-cal resources needed to perform agricultural operations on open ground is considered. The com-bined use of heterogeneous ground and air robotic means expands the functional and sensory capabilities of automatic processing of agricultural land. In a number of cases, for example, when servicing energy supply systems and transporting air means, the problem of physical interaction arises between an unmanned aerial vehicle and a ground service robotic platform. The complexity of solving this problem is associated with the problems of landing, fixing and mechanized pro-cessing of batteries and agricultural resources placed on an aircraft on a service platform, as well as managing the sequence of service of a UAV group. Compared with ground equipment, the use of UAVs in agricultural tasks provides a number of advantages: lack of physical contact with the ground and soil compaction, a wider monitoring and processing area, better treatment of crops with liquid means due to the rotation of rotors without the use of additional devices. Available prototypes of service robotic platforms are distinguished by the complexity of internal mecha-nisms, the speed of service, the algorithms for the joint operation of the platform and the aircraft during landing and maintenance of battery. Autonomous UAV landing in modern research is con-sidered not only on a fixed site, but also on a mobile platform that carries out movement in various environments. Based on the results of the analysis of existing approaches, a classification of exist-ing service systems installed on robotic and mechanized platforms has been compiled. The pro-cessing characteristics of some common crops are considered. A list of operations of the agricultural production process, their duration and cost, as well as the possibility of mechanization. It is concluded that the cost of non-mechanized operations is much higher. A method has been devel-oped for assessing the required composition and quantity of equipment for cultivating agricultural land, characterized by a multi-criteria assessment using a linear combination of three main crite-ria for the total processing time, energy consumed, the cost of the equipment involved and providing numerical modeling and optimization of the volume of involved heterogeneous robotic systems. The results of numerical and simulation modeling of the amount of robotic equipment required for processing agricultural land using arbitrary units and approximate ranges of input parameter values are presented. The simulation was performed in the developed AgrobotModeling software, which also implements visualization of the interaction of unmanned aerial vehicles with agricultural ground service platforms and provides decision support on the optimal number of robotic tools needed to process a given agricultural land area.

When carrying out underwater technical works, walking machines and robots moving along the bottom significantly surpass traditional vehicles in terms of traction and cross-country ability. The exploitation conditions of underwater walking robotic systems - a rough terrain of the seabed, slopes, weakly bearing soil, etc., determine the urgency of the problem of their stability. The paperdiscusses the results of theoretical and experimental studies aimed at ensuring the dynamic stabil-ity of walking machines and robots in underwater conditions. The novelty of the study is provided by the specific features of their operating conditions. The studies are based on the test results of the 6-legged underwater walking device MAK-1. The instability of the walking device may be caused by the gait features. Also, the loss of stability of the walking device can occur when meet-ing with an unrecognized obstacle and when overcoming slopes. Mathematical modeling of the dynamics of statically unstable gait is carried out. The main stages of the phase of movement of the device in an unstable position are analyzed. It is shown that underwater conditions, dynami-cally stable walking of a 6-legged walking device with cyclic movers is also possible in the case of independent legs drives of the right and left side. The methods of autonomous response to a meet-ing with an unrecognized obstacle are considered. Various typical situations that arise when mov-ing along an unorganized surface are analyzed. The methods of self-adaptation and self-management of legs based on fuzzy algorithms, that exclude the occurrence of emergency situa-tions, including rollover, are proposed. The features of overcoming slopes by walking vehicles in underwater conditions are considered. In traditional vehicles movement, rollover or slide downhill is possible. It is shown that on weak soils downhill slide of walking machine is unlikely. This is due to significant deformations of the soil under the supporting elements (feet) of walking machines. A method of increasing the resistance to tipping over when moving a walking device along a slope due to the separate tuning of the conditional clearance of the walking mechanisms is considered. Particular attention is paid to the stability of drilling walking platforms moving along the bottom. Their specificity is the high location of the center of mass. Possible stages of walking platforms rollover are considered. The stabilizing effect of an increased location of the center of floatation is shown. The results of the work can be demanded in the development of walking machines and robots designed for underwater technical works, for new industrial technologies for the develop-ment of the seabed resources, to ensure the antiterrorist and technological safety of underwater infrastructure objects and other works.

The aim of the study is to build human-machine control systems. The main methods for con-structing such systems, methods for isolating evoked potentials in electroencephalograms. The article presents studies of electroencephalogram signals with steady state visual evoked potentials for different frequencies of photostimulation, based on the method of multivariate synchronization index. The influence of the length of the processed window on the accuracy of recognition of the frequency of the studied signal is considered. In the course of research, the authors verify the need for pre-processing of the original signals by means of bandpass signal filtering. In addition, the possibility of using a multi-dimensional synchronization index in multi-channel mode is being considered. The result of the authors study is recommendations on the parameters used to high-light the established visual evoked potentials in the method of multivariate synchronization index. The possibility of using algorithms based on a multivariate synchronization index in real time is shown. The results obtained are of practical importance, since they can be used to build neurocomputer interfaces based on visual evoked potentials and can be further used in the for-mation of control theory of robotic systems for various purposes and in the implementation of solutions for the organization of human-machine interaction in narrow practical problems.

The aim of the study is to increase the efficiency of operation of underwater vehicles (UVs) by using systems of functional diagnosis of their thrusters, by providing detection, localization and identification of minor faults. To solve this problem, the article proposes a new method containing two main stages. At the first stage, a bank of diagnostic observers is built to detect and localize emerging faults. At the same time, each observer is constructed according to a special procedure in such a way to be sensitive to different set of possible faults. At the second stage, additional ob-servers working in the sliding mode are synthesized to accurately estimate the fault values. At the same time, in contrast to existing solutions, it is proposed to use a reduced (having a smaller di-mension) model of the original system when constructing these sliding mode observers. This makes it possible to reduce the complexity of the obtained observers in comparison with the known methods, where full-order observers are built. The results of the research showed the efficiency and high quality of all synthesized observers. In all the considered cases, it was possible to detect the occurrence of typical faults, as well as to ensure the identification of their values. Highly reliable UV control systems can be created on the basis of the considered method.

This work is dedicated to increasing mobile robots autonomy in cases of connection loss with a remote operator. As a number of mobile applications increases, the importance of this task is grow-ing. A reliable way is required to save a robot in cases of communication blackout while operating in a hazardous environment, for example, during a search and rescue operation. Communication blackout means the immediate loss of a robot if it is employed in a dangerous to humans environ-ment. Communication blackouts occur because of communication technologies imperfection, sudden changes in the environment or a human factor. This problem can occur with both wired and wireless communications between an operator and a robot. Therefore, a robot must be able to operate auton-omously in cases operator direct control is lost. A robot must be capable to detect a communication loss with an operator and return to its starting point of a path without human intervention. In this paper, we present developed algorithms for automatic detection of a network connection blackout and autonomous return of the robot. Unlike existing solutions, the developed algorithm does not re-quire additional equipment or software on the operator's side. The robot’s network connection blackout detection algorithm uses TCP / IP packet analysis, which makes it universal for robots con-trolled over Wi-Fi networks. Simultaneous localization and mapping (SLAM) methods and path planning algorithms are used for autonomous robot return. In the autonomous return mode, the robot relies on sensory data collected during movement under the operator control. The algorithms were integrated into the control system of a real wheeled robot PMB-2 and tested in laboratory conditions, which experimentally confirmed their effectiveness and practical applicability.

The aim of the study is to develop a method for controlling the sending of an artillery shell to the gun’s chamber using an acoustic portrait. The existing method for controlling the delivery of artillery ammunition to the gun’s chamber with a separate loading method is based on measur-ing the speed of one of the elements of the rammer. Such an approach to the control didn’t provideguaranteed reliability due to the impossibility of measuring speed in the final segment of the pro-jectile’s inertia motion. At present, vibroacoustic methods of analysis are widely used in various fields of science and technology and can be extended to the problem under consideration. The essence of the method proposed in the article is to excite acoustic vibrations in the "projectile - gun chamber" system and to distinguish characteristic acoustic portraits (signatures) with their subsequent analysis. To study this method, a laboratory bench has been developed that imitates the barrel of a gun with a chamber, and a shell simulator with various obturator belts. A projectile impact at the moment of jamming in the chamber cone or applied externally, for example, on the gun’s body, excites characteristic acoustic vibrations, which differ for cases of reliable and insuf-ficient sending. In the developed stand, acoustic vibrations were excited by an external impact on the resonator and were recorded for subsequent analysis. For an unambiguous classification of events of reliable jamming and insufficient submission, it is necessary to select the optimal vector of signs of an acoustic portrait of the obtained audio recordings. The usual spectral conversion makes it possible to distinguish characteristic frequencies, however, the set of such spectral com-ponents is not suitable as classification features due to the significant array of data obtained as a result of this analysis, and also due to the inability of the Fourier transform to recognize short-term low-power bursts. Therefore, as the classification features were selected mel-frequency cepstral coefficients. Based on the set of such coefficients, using the artificial neural network, the degree of jamming of the projectile simulator in the stand was classified into three categories: “sleep-row is not jammed”, “insufficient shell projectile”, “projectile jammed”. As a result of training the neural network on a significant sample of audio recordings, a classification accuracy of 90% was achieved. It is shown that the developed method of such vibroacoustic analysis can be applied in robotic artillery weapon control systems, as well as in other technical tasks, for exam-ple, in oil and gas production to control the docking of articulated main pipes.

The aim of the study is to assess advantages and disadvantages of existing methods for con-trolling thrust of main propulsions (MP) of unmanned underwater vehicles (UUV). The mathemat-ical model of the MP developed by IMTP FEB RAS was adopted as the object of study. It’s consist-ing of a set of models of an electric motor, propeller and thruster control unit. During the research the following tasks were solved: development of the mathematical model of a brushless motor with refine parameters based on results of its load tests; development of the mathematical model of a propeller based on its action curves determined in accordance with the PROPS model test regres-sion base; development of the mathematical model of an thruster control unit (TCU); simulation of reaction of the thruster for stepwise change of desired thrust with the open-loop regulation of elec-tromotive torque, with feedback on the frequency of rotation and on measured thrust. As the result of simulation main propulsion reaction on stepwise change of desired thrust in bollard pull mode it has been established that different types of thrust control are only differed in transient response time and static control error is almost non-existent for all types of control. Herewith, twofold de-crease in transient response time with thrust and frequency control was found over torque control. This is due to increased power consumption of the motor in the transition process. Modeling of the MP control at the counter flow caused by the movement of the underwater vehicle showed that the control with thrust feedback has the minimum static error and transient response time is compara-ble with the speed control.

The aim of the study is the development of combined control systems by an autonomous un-derwater vehicle (AUV) of the intervention class and a manipulator complex installed on the AUV. Appliances of these types are an important component of underwater resident systems that allow you to expand the range of tasks performed by typical AUVs. The system is multi-level in nature, which allows to properly describe certain states and behavior of the device, depending on the task. The developed system is distinguished by its versatility and modularity, which implies the speed in setting up / adjusting the current tasks assigned to the AUV and the easy implementation and formation of additional tasks by the operator. As an example, we consider the problem associ-ated with the typical work of a resident AUV – sampling of soil fractions. The above results from field tests confirm the workability of the proposed approaches to control, taking into account ex-ternal non-deterministic perturbations of constant and monoharmonic effects. As part of the task, the process of forming the AUV states, transition commands between states, and the formation of a tree of device behavior is described. The scientific and practical novelty of the results presented in the article consists in the implementation of the first developed hardware-software complex for the AUV in the Russian Federation, which allows for the process of soil sampling in both automated and autonomous control modes.

The article considers the features of organizational - technical systems (SOTS), belonging to the class of complex. SOTS may include robotic complexes, automated production, electronic sys-tems and devices used to transmit and convert information, etc. In the modern sense, SOTS are information and technical systems and are currently not only technical objects, they can also be classified as cyberphysical systems (CPS). The effectiveness of SOTS can be determined by many criteria, which should vary in content and time depending on the goals, stage of the existence of SOTS, the influence of internal and external environment. This determines the specificity of mana-gerial decision-making processes in them, requiring preliminary simulation modeling, especially at the design stages of these systems. The general formulation of the simulation problem is given, based on the combination of three approaches to the solution: cognitive, multi-criteria and multi-stage, probabilistic uncertainty. Which are combined into a single complex. Models of a multi-stage decision-making process, a probabilistic model problem of the nominal optimum and cogni-tive modeling of complex systems are proposed A demo example is presented, consisting of the development of a cognitive map of conditional SOTS, functioning in the presence of threats, modeling of functioning scenarios on a cognitive map with hypothetical changes in control and dis-turbing influences on the system. It is shown that at certain stages of decision-making with varia-tions of criteria-based assessments and control actions, it is possible to suppress threats to the system, as well as increase its effectiveness. Simulation was performed using the author's software system CMLS. The developed mathematical and software are designed for intelligent control sys-tems for the rational behavior of complex objects.

The article explains reasons for using computer simulation for mobility enhancement of mo-bile robot systems. The authors focus on the study of the chassis with adaptable configuration of the tracks with the use of the multi-body dynamics software. This approach is asserted to be the most suitable for assessing the cross-country ability and selection of the required characteristics of the drive of vehicles with unconventional running gear. The article also discusses the driving simulator developed at BMSTU for studying the dynamics of mobile robot systems and remotely controlled vehicles. The simulator is based on a mathematical model of the vehicle motion adapted for performing calculations in real time mode. The paper presents a brief description of the model and its main differential equations. In addition to this, a necessary component of the simulator is the software that provides the synthesis of the driving routes based on the known pre-defined sta-tistical characteristics. Thus, the developed simulator allows solving the following main problems: calculating the attainable speed of vehicles and obtaining the load characteristics of the traction drive, as well as debugging control algorithms and studying the interaction of the driver-operator with the vehicle and the environment in the conditions of control signal delay and interference. The simulation method also has an important application in the field of improving the motor-transmission systems of robot systems and remotely controlled machines. Due to the fact that at the moment the most common transmission layout of such vehicles is an individual drive of the driving wheels, increasing the speed of movement of mobile robots requires the use of more pow-erful, and therefore more expensive and heavy electric motors. Thus, the article proposes an ap-proach based on simulation and full-scale and mathematical modeling, which allows to collect the necessary statistical data about the loading modes of the vehicles and determine the required characteristics of electric motors in short-term and long-term operation modes, as well as the desired operating range of the maximum efficiency.

The method of contactless underwater transmission of electricity to an autonomous under-water vehicle (AUV) for charging its rechargeable batteries has practically no alternative when organizing long underwater operation of the AUV. In such system, transformer with separated primary and secondary parts, which are separated by a constructive non-magnetic gap, deter-mined by the total thickness of the connecting walls, is used. The design and electromagnetic characteristics of the transformer significantly affect the overall system efficiency when perform-ing the main task - battery charge for a given time. The presence of a non-magnetic gap requires a special calculation procedure, for which the initial data should be the values of the AUV position-ing errors when underwater approaching the base, along with the parameters of the batteries charge process. The study task is the justification and development of a methodology for calculat-ing the design transformer parameters that meets the specified operating conditions. The research is based on mathematical modeling of electromagnetic processes in a transformer in the ANSYS Maxwell software package in combination with a full-scale experiment. The determining parame-ters in the form of the magnetic coupling coefficient and the winding relative magnetic permeabil-ity are founded; their use is justified for the complete identification of the studied transformer properties. A system of relative units is proposed in which the characterizing parameters have a constant value for cores connected by certain geometric relationships, which makes it easy to scale the results of the obtained technical solutions when changing the requirements for the transmitted electric power. As a result of research, a method for calculating the main design pa-rameters of transformers is proposed. The initial data in the calculation is taken as a combination of the required transformer electrical characteristics and its geometric relations when the re-quired restrictions are met for the permissible error of AUV automatic mooring to the dock-station. The results obtained apply to transformers with cup ferrite cores and with flat magnetic cores, which make it possible to form magnetic cores of the required sizes and configurations. The result of the calculation is the design of the transformer with electrical characteristics that best suit the conditions of its application. Experimental field studies convincingly confirm the correct-ness of the calculation method.

The aim of the research is to develop a complex of mathematical models that provide initial data for a mathematical model of the hybrid energy supply system for subsequent integration into the stand for debugging and maintenance. The work is a development of the previously published mathematical model of the functioning of the hybrid energy supply system of an autonomous un-derwater vehicle. In the work, based on the results of the analysis of the goals and objectives of modeling, mathematical models of electric power sources — a storage battery and an electro-chemical generator — are developed. Since control over the operating parameters of the battery and the electrochemical generator depends on the parameters of the vehicle’s movement, addi-tional mathematical models of the marching propulsion engine and the integrated control system of the vehicle have been developed. The external conditions for the functioning of the vehicle and the route task were set in a specially developed tactical situation simulator. Based on the theory of integrated hierarchical modeling with variable resolution, the most appropriate degree of detail of the developed mathematical models was determined. In view of the need to take into account the non-uniformity of gas blowing of fuel elements in an electrochemical generator, the mathematical model is based on solving a non-linear system of equations, including the Navier-Stokes equation, equations of conservation of momentum, energy and charge. When developing a mathematical model of the battery, the uneven charge of individual batteries was taken into account; The math-ematical model took into account the parameters of individual batteries according to their manu-facturer. The simulation results were the charge-discharge characteristics of the battery. In the mathematical model of the main consumer of electricity - the marching propulsion - the depend-ence of the generated thrust on the required speed of the vehicle is implemented, which allowed to obtain the amount of electricity consumed by the marching propulsion. In the mathematical model of an integrated control system, depending on the current position of the vehicle, motion control-lers are implemented to form control elements of the propulsion system, providing typical modes of maneuvering the vehicle. In addition, the control of the functioning parameters of the hybrid ener-gy supply system was implemented - switching of electric power sources, switching of battery charge processes. In the mathematical model of a tactical situation simulator, the possibilities of defining a route and external conditions are realized. In addition, a model of the vehicle movement was implemented taking into account the forces and moments acting on the vehicle. The developed complex of mathematical models, which provides the data with a mathematical model of the func-tioning of the hybrid energy supply system, can be used as a part of the stand for debugging and maintenance of an autonomous underwater vehicle.

The article explores modern neural network architectures for the automatic detection and recognition of marine surface objects and obstacles of given classes throughout the full image area, applicable for execution in real or near real time on an optoelectronic vision system to au-tomate and improve the safety of civil marine navigation. A formal statement of the problem of automatic detection of objects on images is given. The state-of-the-art algorithms for detecting objects in images based on use of artificial convolutional neural networks were reviewed, their comparison was made and a reasonable choice was made in favor of the most efficient neuralnetwork architecture in terms of computational complexity to recognition accuracy. The subject area is studied, as well as publicly available databases of surface objects suitable for use in the training of algorithms using artificial neural networks. The article concluded that there is insuffi-cient labeled data for training neural network algorithms, as a result of which the authors inde-pendently collected research images and video sequences, prepared and labeled the collected data containing surface marine objects and other obstacles that represent a navigation hazard for ships. Based on the selected neural network architecture, a new neural network algorithm for automatic full-frame detection and recognition of surface objects was developed, and an artificial neural network was trained using the prepared database of images of typical objects. The resulting algorithm was tested by the authors on a validation data set, the quality of its work was estimated using various metrics, and the algorithm’s performance was measured. Conclusions are made about the necessity to expand the collected database of images of typical marine objects, further steps are proposed to improve the accuracy of the developed software and algorithmic complex and its implementation to be used in a marine optoelectronic machine vision system for automa-tion and improving the safety of civil navigation.

In this paper, we consider the main problems associated with the need to integrate various robotic components into a single system while increasing the complexity of the navigation algo-rithms of mobile robots. The work aims to present a hierarchical modular architecture for reconfigurable mobile robots as a solution to the problems posed. In this architecture, a mobile robot is considered as a combination of modules, which in turn consist of simpler blocks - submodules. Each submodule includes a low-power microcontroller and is responsible only for the basic func-tions. A set of submodules forms a module - a transport platform, a robot leg, a manipulator, etc. Besides, one of the main objectives of the project is to provide a framework based on this architec-ture for rapid prototyping of robots from unified modules. The article describes the manufactured prototypes of the modules, briefly discusses the protocol of intermodular interaction of submodules connected by a CAN bus. The results of experiments on testing the protocol are presented and their analysis is given. The efficiency of the proposed solution limitations and a short plan of fur-ther actions for the implementation of the project are shown.

In order to develop a robust algorithm for the automatic detection and tracking of non-deterministic objects for embedded computing systems, in this work, a study and analysis in the field of state-of-the-art general-purpose automatic tracking algorithms is performed. The most successful of those algorithms suitable for long-term stable automatic tracking of objects (without a priori knowledge of the type of object being tracked) have already gone beyond solving exclu-sively tracking problems, and include a synergistic combination of several heterogeneous tracking algorithms, as well as at least one automatic detection and / or classification algorithm. Thus, the authors of the article conclude that the most stable modern automatic tracking algorithms are a multi-agent system that makes a decision about the current position, size and other parameters of the tracked object image based on intelligent voting of system’s submodules that independently monitor the object and form its model. Individual models of each of the submodules are updated based on the results of a collective decision. The authors of the study identified the most effective of the applied basic algorithms suitable for use in embedded computing systems of robotic systems, and developed a new multi-agent algorithm for the automatic detection and tracking of non-deterministic objects. The presented multi-agent algorithm includes a submodule for extracting and matching key points in images, a clustering and filtering submodule for key points using the DBSCAN algorithm, a tracking submodule based on the optical flow calculation algorithm, and a key point classification submodule. A semi-natural testing of the developed algorithm was carried out and its effectiveness in solving tasks not only of automatic tracking of objects, but also in tasks of automatic objects detection using several reference images were evaluated. In conclusion, the authors present steps for further improving the accuracy and performance of the developed algo-rithm for its forthcoming implementation for on-board computing systems of aerial vehicles.

The work is devoted to a review of existing metrics for assessing the quality of the task of tracking objects on video with various algorithms. When evaluating tracking algorithms for their subsequent com-parison, it is not enough to use one metric, and algorithms should be evaluated using a set of different independent estimates. To this end, a study was conducted of existing metrics for evaluating algorithms, the results of which are given in the article. The review involves many different approaches to evaluating algorithms. For example, approaches based on the assessment of the definition of the center of the track-ing object, which are one of the first and still popular metrics for evaluating tracking algorithms. The main disadvantages of such approaches include the difficulty of determining the true center of the object, as well as the interpretation of estimates for various sizes of the object. To eliminate these shortcomings, anew metric is introduced in the article: an unbiased (window) error in determining the center of an object, which takes into account the constant component of the error in determining the center. Other approach-es include metrics based on the analysis of the intersection over union. Also, the article considers ap-proaches based on the analysis of tracking failures, which take into account the tracking length and fail-ure rate. A new method is proposed for evaluating algorithms in case of loss of visual contact with an tracking object, taking into account the number of frames in which visual contact with the object was lost. During the study, approaches to evaluating algorithms for simultaneous tracking of several objects were considered. Integral metrics were proposed whose task is to obtain a comprehensive assessment of the tracking algorithm. For the formation of a comprehensive assessment, it is desirable to use various un-correlated metrics. Complex estimates provide the ability to compare algorithms with each other. As a comprehensive assessment, the article proposes the use of a metric combining the accuracy and robust-ness of the algorithm. As a rule, the intersection over union is used as the accuracy metric, however, for problems where the accuracy of tracking the center of the object is fundamental, the authors propose using an unbiased error in determining the center as the accuracy metric.

Research objective is increase of autonomous mobile means information support efficiency with use of visual data and technologies of its development due to rational use of nonconventional calculators and special preparation of algorithmic providing. Use of nonconventional, heteroge-neous computing means allows to expand significantly a circle of visual data processing problems on the real time scale and, thereby, to increase situational awareness of autonomous robots and efficiency of its receiving. Rational use of nonconventional calculators demands essential altera-tion of algorithmic providing. The majority of visual data algorithms were developed counting on realization on traditional, von Neumann architecture of calculators and demand essential efforts for realization on parallel structures and developments of special programming tools. In work the emphasis on researches in the area use FPGA is placed and a number of approaches in special preparation of necessary algorithmic is considered. As model, demonstration examples of use of nonconventional calculators realization of such algorithms of visual data processing which areactively used in a wide range of information support problems of purposeful movements of mobile robot is considered. Preparation and realization on FPGA of such algorithms as creation of histo-grams, an optical flow calculation, segmentation of images is described. Results of experiments are given in the operating models of airborne computers. As basic data the visual data collected at the movement of mobile means in the conditions of habitat are used. All used software is executed on the basis of the unified software framework of the real time vision systems of domestic devel-opment. In the conclusion further steps in the specified direction, taking into account aspiration to use of domestic software and hardware are discussed.