Activity Recognition and Coordination Analysis of Two-Body Interactions Using Wearable Sensors

This paper describes the application for activity recognition of two-body interactions by using body sensor networks (BSN) and presents a model to give coordination analysis of their cooperation. A monitoring platform based on BSN is established and acceleration data is collected from wearable inertial sensors. Each individual’s signal is divided into several segments, and then recognized corresponding to different stages of movements. The stages’ synchronization between two people could be considered as the principal criteria to evaluate their cooperation quality. Several important performance parameters are also given to provide the necessary feedback for these activities in our system. A gait experiment of two-body interactions was performed on the sensor monitoring platform and the experimental results demonstrate the effectiveness of the proposed method.