Body inertial-sensing network platform for wearable 3D gesture analysis

Abstract

Gesture analysis was widely used in many applications such as healthcare, robotics and human-computer interactions. This paper presented a low-cost body inertial-sensing network platform developed by us. The platform contains the base station, the BSN inertial measurement nodes and the wireless communication protocol. The sensing nodes contain one 3-axis accelerometer, one 3-axis magnetometer, and one 3-axis gyroscope. Wearable gesture analysis was achieved using this platform. Then Kalman filter was designed to get optimal gesture estimation from the BSN inertial measurement nodes. Preliminary results showed that the averaged estimating errors of the roll angle, the yaw angle and the pitch angle were 3.5°, 3.2°, 2.1°, respectively.