IEEE Robotics & Automation Magazine - December 2018 - 32

different system components and
their interplay.
The main lessons learned from the
field trials are as follows:
1) ROV operation from an onshore
control center via a satellite link is
feasible, in principle, despite laten-
cies and low bandwidth.
2) It is, however, important that the
system is capable of detecting the
operator's intentions and using this
information to (semi)autonomously
carry out tasks.
3) There is no black or white with
respect to autonomy, but there are
different levels that can be useful, or
even necessary, depending on the
communication conditions.
4) Good situational awareness through
constant update of the onshore envi-
ronment model is important.

(a)
1
0.8

Simulation

Relative Occurrence

0.6
0.4
0.2
0
0.2

C3
B1
B2
B3
B4

0.4
0.6
0.8

Real Data

90

80

70

60

50

40

30

0
10
20

-9

0
-8
0
-7
0
-6
0
-5
0
-4
0
-3
0
-2
0
-1
0

1
Difference to Ground Truth (°)
(b)
Figure 10. (a) As the basis for (semi)autonomous manipulation, the orientations of the
valves on the mockup structure are determined by an active contours method using
superellipse fitting in combination with a Hough transform. (b) The errors in a single
frame state detection are within a few degrees, and they behave very similarly in the
sea trials (real data) as in the high-fidelity simulation of the system (simulation).

orientations are determined with a Hough transform to
estimate the predominate edges within the fitted ellips-
es (Figure 10).
Conclusions
An approach to underwater manipulation was presented that
facilitates the use of a distant onshore control center with an
exoskeleton based on the following:
1) efficient transmission of multiple data streams over a satel-
lite link
2) a CE to mitigate communication latencies by encoding sta-
tistical models of manipulation tasks
3) the vehicle control, which is more oriented toward AUV
than ROV operations
4) an intelligent vision system, which provides perception
capabilities.
The approach was tested in July 2017 in a first field cam-
paign over two weeks in the Mediterranean Sea near
Marseille. Seven extended dives with about 11 h of
experimental data were performed, wherein the ROV
interacted with a mockup panel structure to validate the
32

*

IEEE ROBOTICS & AUTOMATION MAGAZINE

*

december 2018

Acknowledgments
The presented research was carried
out in the project Effective Dexterous
ROV Operations in Presence of Com-
munications Latencies, which is fund-
ed by the European Commission's
Horizon 2020 Framework Program
for Research and Innovation, under
the topic "Blue Growth: Unlocking the
Potential of Seas and Oceans" and
BG6-2014 "Delivering the Subsea
Technologies for New Services at Sea."

References
[1] P. Letier, E. Motard, and J. P. Verschueren, "Exostation: Haptic
exoskeleton-based control station," in Proc. 2010 IEEE Int. Conf.
Robotics and Automation, 2010, pp. 1840-1845.
[2] G. D. Cubber, D. Doroftei, Y. Baudoin, D. Serrano, K. Chintamani, R.
Sabino, and S. Ourevitch, "ICARUS: An EU-FP7 project providing
unmanned search and rescue tools," in IROS Workshop on Robots and
Sensors Integration in Future Rescue Information System (ROSIN'12),
2012, pp. 1-5.
[3] M. Karaliopoulos, R. Tafazolli, and B. G. Evans, "Providing differen-
tiated service to tcp flows over bandwidth on demand geostationary
satellite networks," IEEE J. Sel. Areas Commun., vol. 22, no. 2, pp. 333-
347, Feb. 2004.
[4] A. K. Tanwani and S. Calinon, "A generative model for intention rec-
ognition and manipulation assistance in teleoperation," in Proc. IEEE/
RSJ Int. Conf. Intelligent Robots and Systems (IROS), 2017, pp. 43-50.
[5] A. Tanwani and S. Calinon, "Learning robot manipulation tasks
with task-parameterized semitied hidden semi-Markov model," IEEE
Robot. Autom. Lett., vol. 1, no. 1, pp. 235-242, 2016.
[6] S. Calinon, "A tutorial on task-parameterized movement learning
and retrieval," Intell. Serv. Robot., vol. 9, no. 1, pp. 1-29, Jan. 2016.
IEEE Robotics & Automation Magazine - December 2018

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2018
