Mathematical model of an artificial neural network for controlling a robotic complex in extreme conditions

Abstract

Mathematical model of an artificial neural network for controlling a robotic complex in extreme conditions

Tsarkova E.G.

Incoming article date: 21.10.2022

The paper considers a mathematical model of an artificial neural network with a delay in the arguments of the state and control functions, designed to control a robotic complex during emergency rescue operations in extreme conditions. The learning process of the considered artificial neural network is described by the problem of optimal control with a delay. Using the Pontryagin maximum principle and the method of rapid automatic differentiation, a method for solving the resulting optimal control problem has been developed. The results of the work of the software tool, the creation of which uses an algorithm for constructing an approximate optimal control of the problem under consideration.

Keywords: emergency rescue operations, artificial neural network, neurocontrol, robotics, maximum principle, safety