A Study on Artillery Meteorological Updating Using Firing Correction Data and Machine Learning

Key Points

• The aim is to enhance artillery firing accuracy by estimating meteorological data through machine learning models.
• Analyzed discrepancies between initial and corrected firing data under changing meteorological conditions.
• Proposed a neural network model to estimate meteorological parameters during firing corrections.
• Utilized historical firing correction data for model training.
• Evaluated the model's performance in varying conditions.
• Significant improvements in firing accuracy were observed with model predictions compared to traditional methods.
• The model successfully estimated meteorological data in real-time, reducing the impact of lagged updates.
• In scenarios with limited support, the model provided a reliable estimate of critical weather elements.

Abstract