Biomedical Engineering Reference
In-Depth Information
in many scientific domains, including biomechanics, neurosciences, robotics and
computer animation. One of the key points is to extract the most relevant parameters
of human walking according to the specific requirements of a given application. In
this section, we propose a summary of the most popular parameters and give some
methods to retrieve them with various sensors.
Nowadays, there are many possible devices to measure human motion and, con-
sequently, human walking. Most of them are generic but some devices have been
specifically designed for human walking. In virtual reality, capturing the intentions
of the user is necessary in order to navigate properly in the simulated environment or
to animate an avatar. While motion capture devices were very expensive and difficult
to use in the past, it is now possible to use cheap and easy-to-use systems such as
the Wii-mote (product of Nintendo) or the Kinect (product of Microsoft) devices.
Whatever the system, the key issue is to design algorithms to extract the relevant
parameters of the user's gait that enable the system to react realistically.
The first part of this chapter is dedicated to global parameters such as walking
speed, step length, frequency and global walking trajectory. The second part of the
chapter focuses on kinematic and dynamic parameters such as joint angles, torques
and muscle activity. We then conclude and give a summary of the studied parameters
and their potential use in walking in VR.
8.2 Sensing and Interpreting Global Gait Parameters
Human being can be represented by more or less complex models. The simplest one
consists in considering human being as a point which corresponds to his center of
mass. Analyzing human walking with such a model consists in dealing with global
gait parameters such as trajectory, velocity, step length, and frequency. It provides
us with relevant information about the global performance of gait. As described in
Chap.
3
those parameters enable us to associate the real-time performance of the user
with multisensory feedbacks, such as adapting the movement of the virtual camera
according to velocity when navigating in virtual environments [
35
]. In this section,
we describe how these parameters are defined and measured.
8.2.1 Step Length and Frequency
When dealing with human walking, one can focus on global parameters such as
step length S
L
and frequency S
F
. These two parameters are in relation with walking
speed V:
V
=
S
L
∗
S
F
As the model is restricted to the user's center of mass, walking speed can be approx-
imated by integrating the signal delivered by an accelerometer placed on the pelvis
