COSMO - AR - testing adaptivity implementations in a factory setting: assemblying a first aid kit

Autor: Vanneste, Pieter, Debeer, Dries, Pinos Ullauri, Luis Alberto
Rok vydání: 2022
Předmět:
DOI: 10.17605/osf.io/ng8vz
Popis: When operators with lower cognitive abilities are trained to perform assembly tasks, Augmented Reality (AR) systems can provide detailed instructional support on-the-job . In this study, we investigate the impact of personalizing the AR instructional support in a real assembly setting, i.e. the composition of a first aid kit in a sheltered workshop. Spatial AR is used through a video-projector, which allows hands-free operation. Two conditions were designed that differ in how the level of instructional support is adapted to individual learner while assembling the first aid kit. There is a non-adaptive condition that shows all operators the same instructional support, for each step and for each repetition, regardless their performance. There is also an adaptive condition, in which the instructional support can vary across repetitions, depending on the performance of the operator. This performance is being estimated through the number of picking mistakes that the operator makes and the amount of help-seeking behaviour. Our hypothesis is that the personalized support (the adaptive condition) has a beneficial impact on objective task performance as well as on subjective learner well being as compared to the non-personalised support (the non-adaptive condition). This study is being conducted within the imec.icon project COSMO, a collaboration between partners from industry (Mariasteen, Rhinox, Azumuta and CNHi), academia (imec research groups Itec, IPI and ID Lab, and Flanders Make) and four Flemish schools. It was co-financed by imec and received project support from Flanders Innovation & Entrepreneurship.
Databáze: OpenAIRE