METHOD OF CONDUCTING BALLISTIC TESTS WITH COMPILATION OF FIRING TABLES Russian patent published in 2023 - IPC F42B35/00 F41F3/04 F41A23/16 

FIELD: weapons; military equipment; testing ammunition.

SUBSTANCE: invention is used in compiling tables of firing artillery guns using machine learning. The method of conducting ballistic tests with the compilation of firing tables for various artillery guns using machine learning methods is that various types of ballistic firing are carried out, the trajectory parameters of the projectile are measured using a complex of testing equipment, the research results are saved in the database in the form of firing tables, ballistics characteristics are calculated and shooting tables are compiled using hardware and software of the range equipment complex. The range equipment complex consists of a module for predicting the characteristics of shooting results, a module for assessing the quality of the forecast, a module for optimizing the program and the scope of research. The module for predicting the characteristics of firing results using machine learning algorithms creates basic predictive models of the first type to predict the characteristics of each ballistic variant of a projectile when fired from a specific type of gun and predictive models of the second type to determine corrections for changes and predict the ballistic characteristics of the results of firing various projectiles from different types guns, which are used as input for new shells and new types of guns. All data on predicting shooting results is entered into the database. The forecast quality assessment module evaluates the current forecast accuracy and the significance of new research to improve the current shooting accuracy. The module for optimizing the program and volume of research adjusts the shooting tables based on the results of assessing the current level of testing and creates a single coordinated database of shooting tables from various sources of information using machine learning algorithms.

EFFECT: creation of a method for compiling firing tables using a machine learning method, which will allow one to analyze firing tables compiled for existing types of guns and ammunition, predict the firing accuracy of new types of guns and ammunition, compile firing tables for them based on the results of a smaller number of ballistic firings, optimize process of planning and conducting tests.

1 cl, 3 dwg