Usability Evaluation Study Of AR Maintenance Induction Teaching System

With the continuous development of Augmented Reality(AR)technology,AR-based aircraft maintenance training has become an important training method.To give full play to its advantages,the usability evaluation of AR training is essential.This paper adopts an AR maintenance induction teaching system based on augmented reality,3D modeling,and other technologies to assist maintenance trainees to better understand and master complex maintenance knowledge and skills,and shorten training time while training more This paper uses AR maintenance induction teaching system to assist maintenance trainees to better understand and master complex maintenance knowledge and skills,shorten training time and train more qualified maintenance personnel.In this paper,the usability evaluation is conducted with the AR maintenance induction system as the research object.The details are as follows:(1)In this paper,the ISO standard is used as the basis,and the usability evaluation indexes are initially determined based on the three-dimensional dynamic selection model of stage,subsystem,and function,and 17 usability evaluation indexes are finally obtained by using the Delphi method to eliminate and correct them.The hierarchical analysis method is used to construct the AR maintenance-induced teaching usability evaluation system and verify that the system is usable.(2)The AR maintenance induction teaching system integrates 3D models with maintenance teaching scenes and provides real-time interactive learning through audio-visual feedback,and the usability evaluation experiment of the AR maintenance induction teaching system based on"EEG+eye movement"was conducted to verify its usability.The experimental group showed high correspondence of graphic trajectories,more red hotspot areas,and even distribution of brain energy in 6Hz,15Hz,and 30Hz regions.The results showed that the experimental group had better experience and concentration than the control group,demonstrating the positive impact of adding cues.(3)Based on the experimental data,17 eye-movement indicators and 5 EEG indicators were obtained.The correlation between the subjective evaluation value and the oculomotor and EEG indexes was analyzed by SPSS software,and an oculomotor model based on"gaze+blink"and an EEG model based on"alpha wave+beta wave+theta wave"were proposed to measure the usability.(4)Based on the Support Vector Machine Regression(SVR)method,the availability prediction model of the AR maintenance-induced teaching system was established by applying MATLAB software,and the coefficients of determination(R~2),root mean square error(RMSE),mean absolute error(MAE)and mean deviation error(MBE)of the training and test sets of the SVR model are positive and small,indicating that the model is highly accurate.And the K-fold cross-validation method was used to evaluate the good performance and verify the feasibility of the model.