Page 323 - Control Theory in Biomedical Engineering

Continuum NasoXplorer manipulator with shape memory actuators 297

different materials that are a mixture of VeroClear and TangoBlack+ in dif-
ferent ratios.
After stiffness tests, it can be observed that there is stiffness change when
the material is heating. There is also some evidence of returning stiffness
when the heating is removed. While comparing across materials, we found
that heating does indeed increase the flexibility of all materials if it does not
break apart. However, constructs consisting of purely one material seem to
yield unstable results. Instead, materials that contain a mix of rigid and flex-
ible materials can yield consistent increases in flexibility while increasing the
heating effect.
This finding allowed us to enhance our device further. By fabricating the
tube in an optimal mix of materials and heat supply, the device is able to
deform into irregular shapes. By doing so, it increases user comfort, as the
device is then be able to account for physiological differences between
patients.

3 Design verification
3.1 Bending capability: Determine the bending angle
A camera was fixed in a position to capture all the videos of the bending
while carrying voltage and pulse-width modulation (PWM). These videos
were passed through an online application of Physlet’s called “Tracker,”
which provided coordinates that could be translated into displacement using
Microsoft Excel, and then converted into a bending angle. A uniform white
background was used in all the videos so that the bending would be more
precise. The device was placed horizontally, as shown in Fig. 6. A small blue
tack was used to lift the NasoXplorer up slightly so that the flexible tip was
not in contact with the paper.
We conducted a bending angle test to determine the highest bending
angle possible for the NasoXplorer (Fig. 7). At 100% PWM, a maximum
angle of close to 80degrees is attainable at 7.5V. However, the large bending
angle of 80degrees over long periods of time may cause fractures to the flex-
ible portion of the NasoXplorer due to the fragile properties of TangoBlack.
Contrary to our expectations, the bending angle for 90% PWM exceeded
that of 100% PWM (a bending angle of 66degrees for NasoXplorer at
100% PWM at 7V). We attributed this to the loosening of the pull wire
in the NasoXplorer, which prevents the efficient transfer of momentum
and force generated by the actuation of the SMA coils. To evaluate the effec-
tiveness of our prototype, we ran various tests to show the optimal