Research On Visual Attention And Cognitive Load In Virtual Maintenance Training Cognition

With the continuous progress and rapid development of science and technology, large-scale equipment structure becomes more and more complex,maintenance technical difficulty and cost are higher and higher. Virtual maintenance training, a kind of method for virtual maintenance using virtual reality technology, solves the problems of traditional maintenance training in the consumption of resources, environmental impact, can’t be analyzed in advance maintenance and other problems, has been widely studied and applied. This paper studies the cognitive load and visual attention of trainees during virtual maintenance training, which includes several parts:(1) This paper introduces the background of virtual maintenance training and the current research situation at home and abroad, summarizes the existing problems and puts forward the main research contents of this paper——the visual attention and cognitive load in virtual maintenance training.(2) The concept of visual attention mechanism and visual attention models,the concept of cognitive load and the classification of cognitive load, the design effect of cognitive load and the methods of cognitive load evaluation are introduced. This part laid the foundation for the following research.(3) Research on the method of reducing the cognitive load in virtual maintenance training. On the basis of studying effects influence the cognitive load, the influence of redundancy effect and channel effect and their mutual effect on the cognitive load of virtual maintenance training trainees under different cognitive load were studied by design contrast experiment.(4) Research on comprehensive evaluation method of cognitive load.Based on the advantages and disadvantages of various evaluation methods of cognitive load, this paper establishes a mathematical model of comprehensive evaluation method, which combines the subjective evaluation method, the main task evaluation method and the physiological index evaluation method, and designs the comparative experiment to verify the reliability of the model.(5) Research on cognitive load evaluation model based on neural network.In this paper, the cognitive load is divided into three grades. The BP neural network model and the probabilistic neural network model are used to evaluate the cognitive load by using the five indicators of subjective, physiological and task performance, and compare the two models at last.(6) Research on visual attention distribution under different cognitive Load. In this paper, we designed several comparative experiment to study the dynamic allocation of visual attentional resources of different trainees within different task environments, and analyzed the effect of different cognitive load on the visual attention distribution of trainees according to task completion.