When constructing the bridge crane control law, a two-pendulum model with a hook suspension, on which the transported load is suspended, and a load suspension is considered. This model more accurately describes the dynamics of cargo movement, which is especially critical in a number of crane operating modes, including cases when the mass-inertial parameters of the cargo are close to the hook. A mathematical model of the two-pendulum mechanical system of the crane suspension along one axis of its movement is described, taking into account the effect of friction forces when moving the crane trolley and wind disturbance acting on the transported cargo. Also, there is presented linearized model of hook movement with control action in the form of preset speed of crane trolley movement. The latter corresponds to the use of a servo motor or stepper motor to control the trolley. On the basis of the obtained linearized model, an adaptive control law is constructed using a scheme including an algorithm for current parametric identification, an implicit reference model, "simplified" adaptability conditions with direct tracking of hook movement (through it — cargo movement) by the properties of the assigned reference model. This control law allows you to build crane control under the current parametric uncertainty of the properties of the crane, the transported cargo and external disturbances (only approximate information about some parameters is assumed). It is shown that in the case when the natural frequency of the assigned reference model is less than that for the hook suspension, the proposed control law generates asymptotic movement of the hook (load) to the assigned point and damping of angular oscillations of the hook suspension and load suspension. For more efficient damping of angular oscillations in case of large values of load moment of inertia, it is proposed to supplement preset value of hook movement with damping movements based on angular movement of load suspension. Model examples supporting theoretical conclusions are given. It is shown that the proposed solution makes it possible to build a bridge crane control under the action of measurement noise and control signal delays.

The relevance of increasing the distribution uniformity of the electroplated coating thickness on the surface of details is considered in the article. The analysis of control actions that affect the uniformity of the electroplated coating thickness was carried out, during which preference was given to the use of multi-anode baths. The existing modes of the electroplated process in multi-anode baths aren’t widely used due to the constructive implementation complexity arising from the need to ensure the uniformity of the resulting coating thickness for multi-assortment industries. The authors propose an electroplated process mode, during which the optimal number and anodes size are found, as well as the current and duration of its flow through them, which ensure the alignment of the electric field lines distribution in the electrolyte between the anodes and the coated detail surface. The control system structure (mathematical model, search algorithm, control and actuator) with feedback on the current through the detail surface is designed to implement considered mode optimal control and an algorithm for its operation (based on the PID law and relay control) is proposed. The practical implementation was carried out using the industrial automation software package CoDeSys V2.3 in the programming language FBD. The proposed mode optimal control was found on the example of applying a zinc coating to a Z-shaped detail. As a result of simulation modeling, the control loop was tuned to the squared deviation integral minimum of the setpoint and the correctness of the control system operation algorithm was demonstrated by a relative average coating thickness deviation up to 5 %, which is comparable to the error of modern thickness gauges.

The thermoelement control system basis on the fuzzy logic has been developed in this article. This system improves the stability characteristics of devices using a PID controller in their structure. This effect is achieved by reducing the output signal, the value of which is calculated using a fuzzy digital filter. The main aspect of the fuzzy digital filter is the use of the area ratio method during defuzzifing. The limitation of this method is the use of triangular membership functions. The experimental results presented in the article demonstrate the effectiveness of the proposed solution. The study has shown that the time of transient processes with the combined control of the thermoelement is reduced by 16—22 %.

A method of swinging of the six-legged robot to ensure its overturn from the "upside down" position is proposed. As a support, we consider an inclined plane with a slight slope towards the flip, with a pit and optionally with a bump next to it. The position of the fixed support can be set by sequential rotations around two different axes. It is shown that the overturn is possible with the help of cyclic movement of the legs, if the body has an upper shell in the form of a truncated cylinder. The legs on the pre-chosen edge of the body through which the flip should occur, are passive, and straightened along the body so that they do not interfere with the flip. The legs on the opposite edge are active; they perform synchronous movement in a plane perpendicular to the longitudinal axis of the body, with a fixed angle in the knee. The results of simulation of the full dynamics of the robot in contact with the support by means of the "Universal mechanism" software package are presented. Specifics of swinging due to the slope, the pit and the bump are shown. For a typical set of robot parameters, the results of numerical experiments for pits of approximately the largest admissible sizes in cases where the support surface is rotated around one axis, around two axes, and for different types of pits, deep and shallow. In all cases, there is a bump next to the pit.

The current stage of development of robotics is increasingly focused on the research, development, production and operation of mobile robots. Their diversity in terms of functional purpose, operating environment (ground, subterranean, subaquatic, aerial, space), according to the requirements for technical, economic and operational characteristics objectively generates a broad variety of such robots. The types of kinematic schemes, propellers and their drives used in mobile robots are also important factors determining their existing and future fleet. The prospects for the implementation and achievement of the aforementioned characteristics in wheeled automatic vehicles (robocars), which constitute a substantial part of modern mobile robots, depend largely on the design of their chassis, and, first of all, on the number and type of wheels used and their relative position on the chassis. Therefore, when choosing a design option for the chassis, the development of robocars involves their multidimensional analysis, based on a comparison of various potentially applicable circuit solutions. The paper contains mathematical descriptions of input-output force-torque mappings (connections) of the shaft torques of the Ilon’s wheels that ensure the motion of robocars along a horizontal plane. The proposed descriptions can be used for comparative analysis and selection of circuit solutions for the chassis of robocars. Such descriptions are specified and analyzed for four variants of kinematic designs of the chassis of robocars with three Ilon’s wheels (Mecanum wheels, Swedish wheels). They are determined for all possible directions of torques applied to the wheels of such robocars, as well as for those cases when no torque is applied to one of these wheels. Thus, the author describes and analyzes the operation modes of such robocars using all three or any two wheels to ensure the forces and torques applied to the chassis of the robocars necessary for their movement. Additionally, the article contains the estimates of the structures of the input-output force-torque mappings for the selected four variants of kinematic designs of the chassis of robocars with three Ilon’s wheels. The influence of their internal input-output cross couplings on the implementation of control algorithms for such robocars is also estimated, as well as the relationship in their controlled variables, maneuverability, controllability and reachability. The author describes a variant of the algorithm for solving the inverse force-torque problems for the chassis in the aforementioned design, and also indicates the preferability for the use of such robocars in the implementation of typical control problems including finite, terminal control, servo and software.

The article is devoted to the justification of group of dynamic systems (DS) operation program based on the failure time prediction. The following assumptions are made: the group includes DS operating under the same conditions, but having different service life; the composition of the DS includes a number of potentially faulty elements with similar values of reliability indicators; for each element a number of parameters are measured and the failure of the DS occurs when at least one of the controlled parameters leaves the tolerance range. The use of a piecewise linear regression model of parameter drift corresponding to the life cycle of DS is justified. The rule of correction of the linear model is formed based on the results of last measurements of the controlled parameters. A unified algorithm for calculating the reliability characteristics in a group of DS based on the drift of controlled parameters is proposed. At the first step, the probability of failure-free operation for a given time and a gamma-percent resource is calculated for one controlled parameter. At the second step, the time moment of a DS failure (which is assumed to happen when at least one controlled parameter leaves the tolerance range) is calculated. At the third step, the reliability characteristics of a group of DS are predicted using a mixture of distributions. The realization of a random variable corresponding to a random mixture is modeled as follows: first, DS is selected at random, then a random variable is modeled in accordance with its distribution function. The mixture distribution function is expressed as a weighted sum of the component distribution functions. Model examples allowing to rank the considered DS according to the predicted failure time with subsequent adjustment of the operation program are considered.