A method for determining the design characteristics of a parallel robot necessary to provide specified parameters of its working space that satisfy the controllability requirement is developed. The experimental verification of the proposed method was carried out using an approximate planar 3-RPR mechanism.

The quality of the diagnostic information obtained in the course of laboratory studies depends on the accuracy of compliance with the regulations for the necessary work. The process of aliquoting blood serum requires immersing the pipette to different depths depending on the boundary level between blood phases. A vision system can be used to determine this depth during automated aliquoting using various algorithms. As part of the work, two recognition algorithms are synthesized, one of which is based on the use of the HSV color palette, the other is based on the convolutional neural network. In the Python language, software systems have been developed that implement the ability of a vision system to recognize blood in test tubes. The developed methods are supposed to be used for aliquoting biosamples using a delta robot in a multirobotic system, which will increase the productivity of ongoing biomedical research through the use of new technical solutions and principles of intelligent robotics. The visualized results of the work of the considered programs are presented and a comparative analysis of the quality of recognition is carried out.

The article considers the task of a tripod robot controlling, which includes two related subtasks: planning the optimal trajectory of motion and its implementation. The realization of the chosen trajectory of motion is carried out by standard methods of control theory and was considered earlier by the authors. Trajectory of the robot’s movement is largely determined by the assigned tasks. Movement of the output link include, firstly, working displacements for the purpose of machining the product and are completely determined by the surface of the workpiece. Secondly - positioning, i.e. the tool movement to the beginning of the next stage of processing. In this case, the motion can be relatively free in some arbitrary trajectory, taking into account the limitations determined by the working area of the mechanism and the surface of the workpiece. For a planar 3RPR robot, the input coordinates are the lengths of the drive links. The output coordinates of the robot (workspace) are limited by the range of permissible lengths of the drive links and the Jacobian sign-sign region. For the obtained working space of the planar 3RPR robot, in the coordinates (1₁, 1₂, 1₃) additional restrictions are introduced, related to the dimensional dimensions of the workpiece. It is shown that the direct use of the Chebyshev metric as an objective function to optimize the positioning path is inexpedient, since Chebyshev’s metric makes a significant ambiguity in the choice of the trajectory. To this end, it is proposed to supplement the original objective function with the Euclidean metric taken with some small weighting coefficient. Optimization was carried out with restrictions on the size of the working space and the workpiece. The results of mathematical modeling are presented.

The article describes the use of a modified ACO algorithm to optimize the trajectory of a planar 3RPR robot. The trajectories of the motion of agents of the first generation were obtained in accordance with the classical and modified algorithm. The efficiency of using the modified trajectory planning algorithm was demonstrated, which made it possible to reduce the length of the original trajectories and the time required to process them. At the same time, in order to save time, the trajectories of movement obtained during its operation can be subjected to discrete local optimization. The result of the work of the algorithms is a set of tasks reflecting the change in each of the input coordinates of the mechanism over time, which can be directly used during the operation of the robot control system. A detailed description of the algorithm and the results of mathematical modeling are presented.

Implementation of a closed-loop control system for a parallel robot is rather difficult due to impossibility of a mechanism output link position direct detection. In order to overcome this, a direct kinematics problem should be solved according to states of the mechanism drive links. However, such approach application requires a solution of a system of nonlinear equations. It is even more complicated for mechanisms with ambiguity of such solution. In this paper, we propose a method to determine the “correct” solution of the direct kinematics problem for such mechanisms using a tripod, a robot with three degrees of freedom, as an example. The method is based on artificial neural networks usage. The data of the mobile platform previous position are used as networks input information. In this paper, it is shown that in this case the adequacy of the obtained neural network based model is sufficiently improved since the neighborhood is defined, where the “correct” solution is located. As a result, both the reliability of the output link position determination and positioning speed and accuracy are improved. The obtained neural network based model can be directly used for the tripod control, and the method to develop this model can be used to synthesize models of other parallel robots.