Research On Evaluation System For UAV Operator Situation Awareness

In recent years,with the development of relevant technologies,the performance and reliability of military unmanned aerial vehicles(UAV)have been continuously improved.However,due to the complex functional interface and the lack of direct physiological connection to the UAV during tasks and training,the reliability of UAV system is reduced,and accidents caused by control errors and decision-making errors in missions is accumulating.It is of great significance to model and measure the cognitive state and situation awareness of operators in UAV operation tasks,either to evaluate the training performance or to select UAV operators.In this paper,the theory of situation awareness and shared situation awareness is discussed,and based on the characteristics of the unmanned aerial vehicle(UAV)operation,put forward on the basis of situation awareness analysis theory of UAV operators situation awareness process,based on the theory of Bolstad,shared situation awareness to understand small UAV operator team situation awareness.In this paper,the direct assessment method of UAV operator's situation awareness and the measurement and evaluation method of situation awareness related elements are proposed,and the feasibility of this method is verified by the experiment of single and two-person UAV operation task.To avoid the draw backs of previous situation awareness evaluation method,this paper introduces Bayesian network and GBDT algorithm into UAV operator situation awareness evaluation task,combined Bayesian networks and GBDT algorithm.In this work,a UAV operator situation awareness evaluation algorithm is proposed.This algorithm uses statistical learning methods to fit a complex function between relevant factors and situation awareness of UAV operators.Based on the direct evaluation method and the evaluation algorithm,we build up an evaluation system of UAV operator's situation awareness.The usability,reliability and stability of the system is tested.Experiment result shows that the proposed system has certain degree of availability and efficiency.