PLANNING THE MOVEMENTS OF ROBOTS BASED ON INTELLECTUAL TECHNOLOGIES

This article is devoted to the issues of planning the actions of robots based on intelligent technologies. One of the main issues in the intelligent planning of robot manipulators' movements is to determine the configuration of the links corresponding to the final position of the manipulator handle in the given Cartesian coordinate. Including the planning of robot movements based on intelligent technologies, the automation of production processes, the discretization of the space of links in the process of planning the movements of robots is important, because this space determines the possible states of the robot handle. The article considers the development of effective methods for reducing the number of possible states and selecting optimal trajectories through space discretization in motion planning of industrial robots. Also, mathematical models such as the Moore-Penrose and Jacobian matrix, which allow determining the current position and orientation of the manipulator handle, minimizing the error between the current state of the robot and the target state, and calculating the inverse problems of kinematics for multi-link industrial manipulators in a universal and stable manner, are considered. Based on the considered mathematical models, an adapted neural network structure for unconstrained inverse kinematics based on the minimization of the goal state error and a neural network-based constrained inverse kinematics calculation algorithm based on intelligent planning of robot manipulator movements are proposed. This neural network structure and constrained inverse kinematics calculation algorithm serve as a basis for industrial robots' motion planning and solving their kinematic problems based on intelligent technologies.