Study On The Relationship Between Controller's Workload And Eye Movement Characteristics

As the air traffic flow increasing,the controller's workload rise up.However,the method of controller's workload measurement is subject,imprecise and difficult to quantify.The air traffic control system is a complex human-computer interaction system.The controller obtains the information of the radar control screen through the eyes.Therefore,the eye movement behavior can reflect the controller's control load and cognitive behavior changes in real time and more accurate throughout the whole cognitive process.At present,the relevant researches mostly focus on a certain aspect of the control and command process,and fail to comprehensively analyze the relationship between the controller's workload and the eye movement characteristics.Based on this,in order to conduct a more comprehensive exploration of the relationship between eye movement characteristics and controller's workload,a series of control scenarios are designed through the construction of experimental platform,and carries out two experiments in succession.On the basis of proving that eye movement characteristics are sensitive to controlling situation's changes,the macroscopic description of the control load is carried out by the eye movement factor.Then,the relationship between the control load and the eye movement characteristics is studied from the airspace structure and the aircraft operating state.The main conclusions of this study are as follows:1.Through clustering the different ways which represent different difficulty of the control situations,namely workload,instruction number,instruction type and eye movement index,and comparing the sample distribution before and after clustering,the result show that the eye movement feature is more sensitive to the change of control situation.2.Selecting the eye movement index which are related to the controller's workload,use the factor analysis method to reduce the dimension,and perform factor analysis to name the factors that can reflect the control cognitive process.The results show that the controller's workload change can be described macroscopically by tracking behavior factor,psychological load change intensity factor,psychological load measurement factor,and the task difficulty measurement factor.3.Dividing the controlled airspace into different regions based on different airspace structures,and the Markov chain model and interval number theory are used to analyze the distribution and transfer of attention resources between these regions in different situations.The results show that the subject spend 80% of attention on the area near the five sides or the areas of complex routs,the load generated by these parts of the areas are also large.4.Dividing aircrafts of different operating states into different AOI(Area of interest),exploring the subjects' cognitive process and the distribution of attention resources to different states of aircraft during the tests.The results show that subjects are highly concerned about aircrafts with potential conflicts,aircrafts closer to the five sides,and aircrafts closer to the conflict zones.These aircrafts generate large workloads.