Industrial Automatic Control Systems and Controllers Annotation << Back Mathematical Model of an Autonomous Object S.M. Ivanova, Z.V. Ilichenkova, A.A. Bukvich The problem of using neural networks to select the optimal path of an autonomous transport object within the design of a control system for technical objects are discussed in the article. Static and dynamically obtained data about the operating environment are used. A method is proposed that allows one to take into account the restrictions on movement: terrain, probable static and dynamic obstacles - and choose the optimal path, taking into account the real dimensions of an autonomous object. An energy functional is proposed that ensures the achievement of a minimum of energy costs for movement under the condition of the safety of an autonomous object. A mathematical model has been built that meets the requirements of safety and optimal movement. The method allows you to find a solution in real time. A neural network providing the choice of a traffic area is presented. Examples of the algorithm’s operation for various operating conditions are given. Keywords: mathematical model; adaptation; autonomous object; neural network; optimal trajectory. DOI: 10.25791/asu.1.2021.1250 Pp. 28-34.