A Survey on Human Activity Recognition using Wearable Sensors

TLDR

Accurate, timely activity and behavior information is crucial in pervasive computing, with applications ranging from medical and security to entertainment and tactical domains. This survey reviews the state of the art in human activity recognition using wearable sensors. The authors outline a general HAR architecture, introduce a two‑level taxonomy of learning approach and response time, discuss key challenges and solutions, and qualitatively evaluate 28 systems on recognition performance, energy consumption, obtrusiveness, and flexibility. The survey identifies several high‑relevance open problems and future research directions.

Abstract

Providing accurate and opportune information on people's activities and behaviors is one of the most important tasks in pervasive computing. Innumerable applications can be visualized, for instance, in medical, security, entertainment, and tactical scenarios. This paper surveys the state of the art in HAR based on wearable sensors. A general architecture is first presented along with a description of the main components of any HAR system. We also propose a two-level taxonomy in accordance to the learning approach (either supervised or semi-supervised) and the response time (either offline or online). Then, the principal issues and challenges are discussed, as well as the main solutions to each one of them. Twenty eight systems are qualitatively evaluated in terms of recognition performance, energy consumption, obtrusiveness, and flexibility, among others. Finally, we present some open problems and ideas that, due to their high relevance, should be addressed in future research.