IEEE Robotics & Automation Magazine - June 2013 - 68

Actuation Results
After the contest, we characterized the propulsion performance of MagPieR. The strategy to achieve the fastest
motion has also been described, and the performance has
been compared with other state-of-the-art propulsion
mechanisms.
The 2-mm dash task at
the NIST and IEEE MMC
The National Institute of
2010 consisted of measurStandards and Technology ing the travel time from
start to complete stop. The
measurement was carried
(NIST) and IEEE initiated
out using a high-speed
an Annual State-of-the-Art camera analysis during the
MagPieR motion from
start to goal line. MagPieR
Microrobotics Challenge,
has several strategies to
boosting the development achieve propulsion efficiency by overcoming surof novel mobile agents with face friction during their
propulsions. First, the
precise and highly dynamic piezoelectric oscillation in
the vertical axis can reduce
surface friction. Second,
propulsion mechanisms
the linearity during its propulsion is assured by pasand controllability.
sively guided motion in
parallel to magnetic field
gradient. Third, the MagPieR is stopped at the closest point to
the goal line.
We chose the MagPieR with the thread crossing the longitudinal axis on the ferromagnetic nickel layer to show the
better-guided linear propulsion. We aimed to reveal the
effect of this thread correlated to the propulsion linearity. For

3
Travel Distance (mm)

impulse. This is a step-by-step actuation, similar to the
stick-slip conditions. Current works are in progress to
achieve useful results for the next NIST challenges.

2.5
2
1.5
1
0.5
0

11 22 57 92 118 135 151171 188 206 223
Input Impulse Step (ms)

Figure 6. Propulsion characteristics of stopping distances depending
on the input pulse steps to MagPieR.

a better precise comparison in propulsion linearity, a highspeed camera with 1,000 frames/s was used to record the
videos during the 2-mm dash task.
Figure 2 shows a series of images at 3-ms interval, captured from the original movie taken at 1,000 frames/s. The
propulsion of MagPieR was revealed to have linearity even
after it passed the goal line. It should be noted here that the
motion of MagPieR was shown in four different distinctive
steps. First, it aligns through the electromagnetic field gradient, continues the linear propulsion, then passes the goal line,
collides to the wall behind the goal line, and finally stops.
This implies that we can further improve the propulsion performance by modifying the input pulse time and controlling
the propulsion linearity from design parameters.
As shown in the velocity plots of MagPieR propulsion, it
collided at around 23 ms (the velocity is zero), which shows
that redundant travel occurs after collision to the wall.
Considering that the travel time estimation for a 2-mm dash
task measures the time between the start line and a complete
stop behind the goal line, this redundant motion can be
avoided to further improve the record by about 25-30%.
To avoid the collision to the wall and make the MagPieR
stop at the closest distance from the goal line, we aimed to
find the optimal input step pulse. For this
purpose, we characterized the travel distances of MagPieR with different input
impulse times. The measured result is
presented in Figure 6. As a result, 14 ms
of input impulse time stops MagPieR the
closest to the goal line. This parameter
could reduce the travel time by avoiding
the redundant travel behind the goal line
from collision and bouncing. Further,
surface optimization can even enhance
10 cm
the current propulsion performance.
Furthermore, the initial alignment is
(a)
(b)
important to achieve the propulsion linearity. When the thread is initially well
Figure 5. (a) The setup CAD schematic showing the coils and the interchangeable
aligned through the field, additional
arena. The arena consists of four layers (from top to bottom): An ITO glass top electrode
motion required for the alignment (a
(1.1 mm), a glass spacer (155 n m), a silicon arena border (110 n m), and, finally, a silicon
base substrate acting as the ground electrode (400 n m). (b) The photo of the setup.
damped oscillatory rotation) that adds

IEEE ROBOTICS & AUTOMATION MAGAZINE

june 2013

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2013