As a result, the measurement of the height of accuracy
about 1/200 is possible for a perpendicular movement.
However, accuracy is very bad when leaning by hand. The
cause of this is thought to be that for the acceleration to
work horizontally in time when the device is lifted from the
level stand, putting, and when had the device by the hand,
and to calculate accelerations other than the direction of
the gravity axis as combined power.
3-2. Horizontal basic experiment
To do the experiment by which the height (error) calculated
when horizontally moving was confirmed, by movement
time regularity, distance was changed and it moved on the
almost horizontal passage of the third floor of the fourth
Chiba Institute of Technology building (Figure 4). Each
distance of the horizontal distance 10m, 15m, 20m, 25m,
30m, and 35m was measured at 30 seconds.
Measurement was performed on the sampling frequency
of 50Hz. Figure 5 shows the result
Figure 4 Experiment place
Error by horizontal acceleration
E20
Sad + $
5 10 © :
Mat $e
3 05 | Ts
©
S 00
o $
ul -0.5
0 10 20 30 40
Horizontal movement distance (m)
Figure 5 Experiment result
280
An experiment result shows that the large error has come
out of a quicker motion. As a result, the influence of the
horizontal acceleration is understood. How this horizontal
acceleration is processed because the horizontal
acceleration greatly influences in the measurement which
uses the combined power of three axes becomes a
problem.
4. APPLICATION TO FIELD EXPERIMENT
4-1. Ideal model
The influence of the horizontal acceleration is considered.
The horizontal acceleration is included in the combined
power of three accelerometers as shown in Figure 6. To
examine the influence of the horizontal acceleration, the
following models were examined theoretically.
Combined power of acceleration a, b, and c
Acceleration c
Acceleration b
Acceleration a
Acceleration of
combined power
Gravity acceleration
-
Vertical acceleration
Horizontal acceleration
Figure 6 Influence of horizontal acceleration
«The conditions of a model»
It goes straight on a horizontal plane, and it is accelerated
to 40km/h (11m/s) at 8 seconds. It runs 40km/h for 85
seconds, and it stops at 5 seconds. The gravity
acceleration is assumed to be 9.8m/s°. On the basis of this,
it verifies by actual motion (Figure 7~10).
4-2. Comparison with Ideal model
It experimented by considering application in an actual
geographical feature from the result of a perpendicular
basic experiment and a horizontal basic experiment, and
carrying measurement device in a car.
Horizontal
CAT nN SOL Ae Ta Rt pn EY
velocity(m/s)
One pie
accelerat
advance
line road
height ab
City, Chit
The mot
km/h)—
The velo:
same tim
speed n
experime
inertial «
accelerai
one wat
Speed (
where e
progress
fixed vel
experime