Semantic Analysis Research Of The Air Traffic Control Phraseology Based On Natural Language Understanding

The air traffic control system manages the takeoff and navigation of multiple aircraft,thecontroled aircraft relys on visual, radio communication and navigation means to performcontrol tasks. Radio communication as a means of communication between the controller andpilot, is an important factor in controlling the safe operation of the aircraft. In order to ensurethat the information between controllers and pilots is accurate, not only for communicationand other hardware equipment to operate with high quality, but also the controller and pilotbusiness skills raised a requirement. With the development of science and technology, theless danger as a result of the equipment and other hardware problems, human factors has ledto the dangerous upward trend in all of the risk factors. Major human factors: poor handlingcapacity in the special case, misallocation of attention, etc. caused that the pilot did notunderstand the controllers intent, the controller can not be correctly received pilot readback,and thus can not determine whether the pilot received correct commands. So the analysis andresearch for ATC phraseology has realistic meaning for ensuring that the construction ofnational aviation safety and the sustainable development of air transportation.The purpose of the study of the semantic analysis of the ATC phraseology is to ensureflight safety and order, reduce controller and pilot workload and work pressure, improvecontrol efficiency. Semantic parsing of control terminology, including segmentation,dictionary,Semantic analysis, command transformation.ATC phraseology has its own rules and characteristics. The issues to consider to establisha phraseology semantic parsing system suitable for ATC conversation. Analysis the languagecharacteristics of ATC conversation language, Analysis dictionary suitable for ATCconversation language, research related keyword matching algorithm and gives excellentshortcomings,according to the actual ATC conversation language design Semantic analysistree and associated system command. Through segmentation, semantic parsing, commandtransformation and combined with the actual ATC conversation language,extract the keycontrol command and flight status information, and converted to normalized systemcommands, make consistency detection for pilot readback,if there is deviation, given warninginformation. Through the analysis of the semantic parsing system, basically the flightinformation and command are extracted, converted to normalized commands, Give a warningfor the error in the pilot readback, further ensure the effectiveness of communication betweencontrollers and pilots.