IEEE Robotics & Automation Magazine - September 2014 - 59

50

100

150
Time (s)
(a)

200

250

y-Position (mm)

0

Still Water Strategy
Desired Trajectory
Flow Strategy

300
250

Flow

200
Trajectory 1
150

Trajectory 2

100

450

300

400

300

0

500
600
700
x-Position (mm)
(a)

320

150
0

50

100

150
Time (s)
(b)

200

250

Figure 9. (a) The orientation and lateral displacement of the
robot recorded with an overview camera [25]. (b) The robot is
held in its longitudinal position with the feedback control from
the camera for recording while the orientation is controlled by
the Braitenberg controller using flow sensors or by the camera
(for comparison). With no heading control, the robot is bouncing
between the flow tank's walls (blue line) while the precision of
the heading control based on flow sensing (red) and camera
feedback (green) are almost comparable.

downstream drift at the end of the experiment was 1/5 of the
robot's body length [25]. Currently, a standard method in
underwater robotics is to use an acoustic Doppler current
profiler (ADCP) to estimate the bulk flow speed. ADCPs are
bulky, costly, and energy-consuming devices that are not
suitable for small underwater vehicles. The lateral-line-based
odometry can here provide a low-cost alternative for estimating the flow-relative speed and providing the velocity estimate in the absence of a global reference.
Experiment 4: Reducing Energy Consumption in
Turbulence by Exploiting Vorticity
The experimental setup of this experiment is different than
that shown in Figure 8 in that the robot is harnessed to a force
plate for force measurements. The lateral line signals are used
to control the tail beat timing. A motion pattern similar to
Kármán gaiting is achieved by adjusting the frequency of the
tail beat to the vortex shedding frequency and fine-tuning the
tail beat timing. The tail bends against the high-pressure zone
created by the vortex, and the robot takes advantage of the
increased perpendicular component of the lift force created by pressure differences on both sides of the flexing tail
(Figure 11). The results are compared with those of the tail fin
propulsion in the steady flow with the same incoming flow
speed. In comparison, 100% more thrust is created in KVS
with the appropriate tail beat timing than in steady flow [30].

800

900

Actual Position
Desired Position
Errors

300 Flow
280
y-Position (mm)

Orientation (°)
Lateral Position (mm)

30
20
10
0
-10
-20
-30

350

No Lateral Control
Camera Feedback
Braitenberg Control

260
240
220
200
180
160
140
520

540

560

580 600 620
x-Position (mm)
(b)

640

660

680

Figure 10. A flow-aided trajectory following [29]. (a) A straight
line in the flow (dashed line) using the sideslipping maneuver
and flow information (thin line) and without flow information
(thick line). (b) A closed trajectory in the flow. Laminar flow is
directed from left to right, and the robot's pose in the flow is
marked in light blue.

Experiment 5: Reducing Energy Consumption by
Holding Station in a Hydrodynamic Shadow
We first identified the reduced flow zone [zone 2 in Figure 5(d)] behind the object from DPIV images and adjusted
the thresholds of pressure readings to identify the point of reference for station-holding. The robot compensates for the lateral and longitudinal drift using PID controllers and keeping
the station in the hydrodynamic shadow. Monitoring the
motor current consumption reveals that swimming in the
reduced flow zone consumes 7% less energy behind the cylindrical object. The standard deviation of the downstream position behind the cylinder is 40.5 mm, and that of the lateral
position is 12.7 mm. When the cylinder was replaced by the
cuboid, creating a sharper pressure drop between the suction
zone and the reduced flow zone, the energy consumption
reduced by 17%. The standard deviation of the downstream
position was 21.2 mm, and that of the lateral position was
13.3 mm in this experiment. The duration of both experiments was 270 s [25]. The trajectories of the robots are plotted in Figure 12.
september 2014

*

IEEE ROBOTICS & AUTOMATION MAGAZINE

*

59
Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - September 2014
