Research Of Radar Dynamic Target Calibration Method Based On Eye Tracking

As an emerging technology,eye tracking has become one of the important research methods in the study of human factors in aviation field.Target calibration is the base of conducting the eye movement research.The target on which air traffic controllers focus is the dynamic target in radar interface.Studying the dynamic target calibration problem of radar interface is of great value for improving the safety level of air traffic control.The Tobii X2-60 eye tracker and radar control simulator were used to build an experimental platform,and eight controlled trainees were recruited to participate in the simulation control experiment.Combining experimental data,starting from the static target calibration research,with the analysis of the distribution characteristics of eye movement data,a radar static target calibration method based on noise-based density-based clustering algorithm(DBSCAN algorithm)and convex hull algorithm is proposed and utilized.The K nearest neighbor algorithm(KNN algorithm)and variable density threshold setting method improve the algorithm and improve the adaptability of the algorithm to eye movement data.At the same time,based on the similarity analysis and coverage calculation,the calibration results of different static targets are compared.The results show that the similarity between the calibration areas of different targets is greater than 80%,and the coverage is greater than 90%.Based on the correlation analysis,the coordinate transformation model is selected to establish the coordinate transformation relationship between the aircraft ground coordinates and the screen coordinates.According to the track information of the aircraft,the screen motion trajectory of the aircraft in the radar interface is obtained.The static target area is translated by using the aircraft screen motion trajectory to achieve calibration of the dynamic target.The proposed dynamic target calibration method is tested by multi-thread synchronous programming.The results show that the average calibration accuracy of the method reaches 81.84%,which has good calibration accuracy.