Design and Analysis of User Studies for the Evaluation of a Human-Robot Interaction Scenario: User Studies
Author
Rodriguez Pinto, Daniela
Term
4. semester
Education
Publication year
2022
Submitted on
2022-01-06
Pages
70
Abstract
Industri 4.0, den hurtige digitale omstilling af produktion, bringer nye teknologier og arbejdsmåder, men kan være svær at indføre. Dette projekt udvikler en Learning, Training, Assistance (LTA)-model og afprøver en sprogunderstøttet virtuel assistent til at støtte medarbejdere. I et brugerstudie samlede deltagerne en mockup-telefon med hjælp fra en industrirobot og en virtuel assistent ved navn Max. Efterfølgende udfyldte de spørgeskemaer om systemet og interaktionen. Systemet opnåede et gennemsnit på 68,53 på System Usability Scale (SUS). Den samlede oplevede frustration havde gennemsnit på 2,41 ved interaktion med Max og 1,74 ved interaktion med robotten. Når deltagerne skulle vælge foretrukken støtteform, valgte 69,0 % Max, 31,0 % en brugermanual og 0,0 % en uerfaren kollega. Deltagerne gav desuden nyttige forslag til at forbedre den nuværende løsning og til fremtidige versioner.
Industry 4.0, the rapid digital transformation of manufacturing, brings new technologies and ways of working but can be hard to adopt. This project develops a Learning, Training, Assistance (LTA) model and tests a language-enabled virtual assistant to support workers. In a user study, participants assembled a mock-up phone with help from an industrial robot and a virtual assistant named Max. Afterwards, they completed questionnaires about the system and the interaction. The system achieved an average System Usability Scale (SUS) score of 68.53. Reported overall frustration averaged 2.41 when interacting with Max and 1.74 with the robot. When asked how they would prefer to get help, 69.0% chose Max, 31.0% chose following a user manual, and 0.0% chose interacting with an inexperienced coworker. Participants also offered practical suggestions to improve the current system and inform future versions.
[This summary has been rewritten with the help of AI based on the project's original abstract]
Keywords
Documents