Multi sensor system for pedestrian tracking and activity recognition in indoor environments

Abstract

The widespread use of mobile devices and the rise of global navigation satellite systems (GNSS) have allowed mobile tracking applications to become very valuable in outdoor environments. However, tracking pedestrians in indoor environments is still a very challenging topic. This paper presents a system for pedestrian tracking and activity recognition in indoor environments using exclusively common off-the-shelf sensors embedded in smartphones. The proposed system combines the detection of biomechanical patterns of the human body while accomplishing basic activities along with identifiable signatures that certain indoor locations introduce on sensing data. The system was implemented and tested on Android-based mobile phones. The system detects and counts steps on a flat surface or stairs, tracks users' changes of direction and altitude, and recognises the proposed human activities with accuracy. All combined this leads to a total tracking accuracy of 91.06% in common human motion indoor displacements.