Phantom: Prediction of Human Motion with Distributed Body Sensors - Prediction and Classification of Respiratory Motion - page 37

Chapter 3
Phantom: Prediction of Human Motion
with Distributed Body Sensors
Tracking human motion with distributed body sensors has the potential to promote
a large number of applications such as health care, medical monitoring, and sports
medicine. In distributed sensory systems, the system architecture and data pro-
cessing cannot perform the expected outcomes because of the limitations of data
association. For the collaborative and complementary applications of motion
tracking (Polhemus Liberty AC magnetic tracker), we propose a distributed sen-
sory system with multi-channel interacting multiple model estimator (MC-IMME).
To figure out interactive relationships among distributed sensors, we used a
Gaussian mixture model (GMM) for clustering. With a collaborative grouping
method based on GMM and expectation-maximization (EM) algorithm for dis-
tributed sensors, we can estimate the interactive relationship of multiple sensor
channels and achieve the efficient target estimation to employ a tracking rela-
tionship within a cluster. Using multiple models with improved control of filter
divergence, the proposed MC-IMME can achieve the efficient estimation of the
measurement as well as the velocity from measured datasets with distributed
sensory data. We have newly developed MC-IMME to improve overall perfor-
mance with a Markov switch probability and a proper grouping method. The
experiment results showed that the prediction overshoot error can be improved in
the average 19.31 % with employing a tracking relationship.
3.1 Introduction
Prediction human motion with distributed body sensors has the potential to
improve the quality of human life and to promote a large number of application
areas such as health care, medical monitoring, and sports medicine [ 1 - 3 ]. The
information provided by distributed body sensors are expected to be more accurate
than the information provided by a single sensor [ 1 , 4 ]. In distributed sensory
systems, however, the system architecture and data processing cannot perform the
expected outcomes because of the limitations of data association [ 5 - 14 ]. As shown
in Fig. 3.1 [ 15 ], individual sensory system using IMME shows the position