Research On Visualization Of NOTAM Based On Natural Language Processing

In recent years,duo to the pilots do not know the working condition of the scene structure,the flight safety problems occur frequently.The excessive number of NOTAMs causes the most important reason of these accidents.In addition,the content of the NOTAM is difficult to understand,and the format is non-standard.However,the CNMS cannot process and display the contents of NOTAMs.In view of this,the natural language processing technology is introduced into the processing of notice of navigation and makes visual display of NOTAMs from the pilot’s point of view to solve the safety problems caused by NOTAMs.First of all,some flight safety problems at home and abroad are analyzed,highlights the shortcomings of the current NOTAM system,integrates the application of NLP and digital map technology in NOTAM processing,and designs the functions of NOTAM visual display system;Then,duo to the content of the NOTAM is difficult to understand for pilot,the paper uses SMT to translate the text of item E of NOTAMs into Chinese,and evaluates the translation effect by BLEU model;Next,the text similarity technology is used to identify the non-standard NOTAMs.By comparing the cosine similarity algorithm,the improved TF-IDF algorithm and the Word2 vec algorithm,the algorithm with the best effect is selected to determine the recognition standard of the nonstandard NOTAMs;Finally,the improved IMM-PF algorithm is introduced to improve the accuracy of tracking aircraft position,and the visual display is realized on the digital map of the airport with the processed standard NOTAMs,which effectively solves the flight safety problems caused by NOTAMs.