Information Technology Reference
In-Depth Information
accelerations and a timestamp. Fig. 11.1 shows three different synchronized pic-
tures. The first one (at the top) illustrates the dorso-ventral acceleration (
a
x
)and
the medio-lateral acceleration (
a
y
) obtained from the three-axial accelerometer. The
middle picture plots a sketch of a person representing the different phases of the
gait with the right limb boldfaced. Finally, the picture at the bottom represents the
time period from one event (usually initial contact) of one foot to the subsequent
occurrence of initial contact of the same foot.
1.5
1
0.5
0
−0.5
a
x
a
y
−1
0
0.2
0.4
0.6
0.8
1
1.2
t (s)
Fig. 11.1
One gait cycle illustrating the various phases and events and the dorso-ventral (
a
x
)
and medio-lateral (
a
y
) accelerations
11.2.2
Fuzzy Finite State Machines
As said, we will consider a FFSM to deal with the human gait modeling problem.
The theoretical basics of FFSMs were established by [13] and later developed by
[6, 10, 18]. The model of FFSM presented is inspired by the concepts of fuzzy
state and fuzzy system developed by Zadeh [19, 22]. More specifically, it can be
