Trajectory Correction of Autonomous Objects in the Presence of Restrictions

The traffic safety of an automatic transport facility is one of the most pressing and interesting problems. The use of neural networks to solve this problem is the most promising development. The problem of constructing the trajectory of a transport object in the event of a dead-end obstacle is described in our article. A model of the control system is proposed, in which the constantly updated information received from the sensors is analyzed. At the same time, it should be noted that this information contradicts some unchanged data that are taken from global sensors. An external object that can change the rules for the movement of vehicles in the current conditions conducts data analysis. Based on the results of this analysis, it is proposed, according to a certain algorithm, to change the route of the autonomous object in accordance with the shortest option for returning to the original trajectory. The article defines a neural network according to the energy function that minimizes the cost of movement and ensures the safety of an autonomous object.