Research On Air Traffic Decision-Making Method Based On Extraction Of Natural Semanteme

The first task of air traffic control is to ensure the safety of the aircrafts in the airspace by dispatching aircrafts flexibly to eliminate potential conflicts rationally,and to ensure the safety of the operation.In today air traffic control system,the flights are based on the flight plans,when ground air traffic controllers get flight information through pilot position reports or radar detection and other ways,and then they would send the appropriate commands to the aircraft according to the situation of flight airspace.The flight commands are included in the aircraft altitude,speed,heading and flight parameters.However,in the area with large air traffic flow or complex airspace,the ATC workload caused by the command pressure of controllers has a negative impact on the command decisions and the issuance of instructions.And the controllers for the upper air cannot monitor the aircraft effectively,there are many uncertainties in ATC.In this case,unreasonable or even unsafe instruction conferences bring great operational risks to the operation of aircraft.Therefore,in certain airspace or airport terminal area with large traffic density,it is very important to adopt a certain method to evaluate the rationality and scientificity of the command and decision.Therefore,the speech recognition system based on the development of Xunfei provides for air traffic control of the scene,the command voice of ATC is stored as text form,using the prefix factor of regular expression matching based on instruction filtering method to extract information from the voice control instruction text;then constructs the aircraft flight speed,combined with the aircraft characteristics of the flight altitude and heading status change model,is presented based on the air traffic situation recognition based on Hidden Markov BP based on neural network,can effectively control the current state of knowledge of flight aircraft not under the instruction of airspace within the airspace operation situation;finally,selection of air traffic situation evaluation index,using grey clustering method to evaluate the air the traffic situation,determine the current air traffic control command.In order to effectively feedback the scientific and reasonableness of the control decision,the situation level is carried out.This paper applies a method to assist ATC's decision-making effectively through the perspective of speech recognition,the recognition,construction and evaluation of air traffic situation.The proposed methods have strong pertinence and are easy to implement.It would improve the air traffic control method of decision support theory.