The Analysis Of Aircraft Taxiing And Prediction Study Based On Surveillance Data

With the rapid increasing of aircrafts, the operation efficiency of airport becomes lower and the taxiing time of aircrafts becomes longer. Meanwhile, the layout of airport is more complex, leading to potential taxiing collision. On this basis, it is very important to study taxiing process of aircrafts. History surveillance data is analyzed to figure out characteristics of taxiing process, including taxiing time, taxiing route, taxiing velocity and so on. The position label of aircrafts to draw a whole process of aircraft taxiing is predicted. The airport hot spots identification methods are discussed and found out by cluster method. The prediction of taxiing time based on history surveillance data is accomplished by use of multiple linear regression method.The result of SVM prediction displays high precision at predicting airport label, which provides important assurance to do further analysis. Two methods are discussed to find airport hot spots, which provides comprehensive ways to find airport hot spots and provides support for taxiing strategy. It shows high precision that using multiple linear regression to predict taxiing time, especially analyzing factors in regression equation.