Comfort Wearables for In-Flight Sitting Posture Recognition

Xinhe Yao, Yusheng Yang*, Gerbera Vledder, Jun Xu, Yu Song, Peter Vink

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

28 Downloads (Pure)

Abstract

Wearables are used to recognize human activities in various applications. However, there is limited evidence on the comfort feelings in using wearables, which is crucial for the adoption and long-term engagement of users in those applications. In this paper, we propose the concept of comfort wearables in the context of in-flight posture recognition. A comfort wearable and a tight-fit version, using identical hardware and software architecture, were prototyped and tested by 35 participants in a Boeing 737 cabin. During the usage of each wearable, participants were asked to perform seven frequently observed in-flight sitting postures and report their overall comfort/discomfort afterwards. A multilayer perceptron neural network was used to classify those activities. Experiment results indicated that participants appreciated the comfort wearable, rating it with significantly higher comfort scores and lower discomfort scores. Cross-validation results also revealed that using the comfort wearable achieved even better accuracy (74.8%) than using the tight-fit wearable (65.8%) in posture recognition. Outcomes of the study demonstrate that ergonomic design and technical accuracy are not competing factors in the wearable design and highlight the opportunities for designing and using comfort wearables in broader contexts.
Original languageEnglish
Pages (from-to)103659 - 103668
Number of pages10
JournalIEEE Access
Volume11
DOIs
Publication statusPublished - 2023

Keywords

  • Losse-fit
  • tight-fit
  • ergonomics
  • accelerometer
  • IMU
  • wearability

Fingerprint

Dive into the research topics of 'Comfort Wearables for In-Flight Sitting Posture Recognition'. Together they form a unique fingerprint.

Cite this