Mirrored Motion

A Wearable, Body-Area Network Wireless Sensor System, for Pervasive Body Motion Capture

Humans have always had difficulties interfacing with computers. The difference in language is perhaps too great to ensure natural and effortless communication. Interaction can be improved by making a computer more human-like. Today many artificial intelligent technologies like speech and image recognition systems are commercially available to give computers more human qualities. This project investigates motion-capture as one such a technology.

Yet motion-capture is not limited to man-machine interfacing only, but also has applications in a diverse field of disciplines. The problem addressed in this study is specifically aimed at the development of a low-cost, real-time motion-capture system. The approach taken here is to use sourceless sensors to establish the orientation of the human anatomical segments, from which posture is then determined. The orientation sensors consist of two three- axial sensing units. Orientation is obtained by combining (or fusing) these information sources into a rotation matrix – an algebraic format that can be directly applied to find the posture of the actor/user.

A simple skeleton model is defined as a coarse representation of the actor. A kinematical approach is used to manipulate the skeleton. Wireless technology is used to connect the sensors to a body area network. Wireless sensors make the system unobtrusive and, compared to a wired solution, allows for a much wider range of applications.

An off-the-shelf network wireless network based upon the IEEE 802.15.4 standard is selected, and proved to fulfil both the real-time and low-cost project requirements.

Our system, built in 2007, has gone through two iterations of the hardware phase, in which we have tried to miniaturise our sensor nodes for a truly pervasive and wearable solution.