Cognitive Load Theory Based Evaluation Method Of Voice Human-Robot Interaction

With the development of robot technology and market,robot has gradually become one of the most important electronic consumer products,and the related interaction design issues are increasingly prominent.In the field of robot voice interaction,due to the great difference between voice interaction and traditional graphical user interface,it is difficult to fully use visual design principles for user experience evaluation of voice interaction.The design methods and evaluation methods of robot voice interaction need to be established.Based on the cognitive load theory,this paper proposed an evaluation scheme of robot voice interaction design,which combines subjective measurement and physiological data verification.Firstly,this paper defines the basic concepts of robot voice interaction research.Based on introduction of the cognitive load theory,the characteristics and levels of robot voice interaction interface were explained,and the basic principles of robot voice interaction design were proposed.Secondly,based on the principles of the Task Load Index(NASA-TLX),a evaluation index system of robot voice interaction cognitive load was summarized by using Analytic Hierarchy Process(AHP),and a set of design methods for robot voice interaction were developed.Through a experiment of prototypes simulating the process of robot speech interaction,the study obtained the specific values of the voice interaction cognitive load evaluation index weight which completed and improved the evaluation system.Thirdly,following the guidance of the experimental results and the improved design methods,the research re-designed sweeping robot as an example for the robot voice interaction design.Through in-depth user study,this article summarized the user scenarios and specific using problems for the classic sweeper robots,and then designed 3 voice interaction systems based the analysis.Finally,this paper studied the combination of subjective cognitive load measurement and heart rate and galvanic skin response(GSR)measurement,so that they can confirm each other and get reliable cognitive load evaluation result,which proves the reliability and practicability of the evaluation system.In conclusion,based on the cognitive load theory,this paper designed 3 voice interaction systems of sweeping robot.By using the basic principles of Analytic Hierarchy Process,the traditional subjective measurement tool of cognitive load was studied and improved.A set of voice interaction cognitive load evaluation system including 4 criteria and 14 indicators was established.The system was verified by a physiological measurement of heart rate and GSR signal which indicated that a new technical means for the research of human voice interaction user experience is obtained.