IEEE TRANSACTIONS ON ROBOTICS AND AUTOMATION, VOL. 20, NO. 3, JUNE 2004 499
Ground–Space Bilateral Teleoperation of
ETS-VII Robot Arm by Direct Bilateral Coupling
Under 7-s Time Delay Condition
Takashi Imaida, Yasuyoshi Yokokohji, Member, IEEE, Toshitsugu Doi, Mitsushige Oda, and
Tsuneo Yoshikawa, Fellow, IEEE
Abstract—A bilateral teleoperation experiment with Engi-
neering Test Satellite 7 (ETS-VII) was conducted on November 22,
1999. Round-trip time for communication between the National
Space Development Agency of Japan ground station and the
ETS-VII was approximately seven seconds. We constructed a
bilateral teleoperator that is stable, even under such a long
time delay. Several experiments, such as slope-tracing task
and peg-in-hole task, were carried out. Task performance was
compared between the bilateral mode and the unilateral mode
with force telemetry data visually displayed on a screen. All tasks
were possible by bilateral control without any visual information.
Experimental results showed that kinesthetic force feedback to
the operator is helpful even under such a long time delay, and
improves the performance of the task.
Index Terms—Bilateral control, ground control, passivity, space
manipulator, teleoperation, time delay.
I. INTRODUCTION
B ILATERAL CONTROL provides important force in-formation on a remote environment to an operator. It is
well known, however, that even small communication delays
may destabilize the system with conventional bilateral control
methods, such as symmetric position servo and force-reflecting
servo [17]. Anderson and Spong [1] proposed a bilateral control
law that maintains stability under communication delays by
using the scattering theory. Niemeyer and Slotine [12] studied
further on this problem.
It has been assumed, however, that bilateral control methods
would not be effective when the time delay becomes longer than
about 1 s. For example, Kim et al. [6], who conducted an exper-
iment of peg-in-hole tasks using a force-reflecting servo under a
time-delay condition, described it as, “… However, this force-re-
flection technique can be utilized only up to an approximately
0.5- to 1-s communication time delay, since a long time delay
in the force feedback loop causes the system to be unstable.”
Manuscript received July 4, 2003. This paper was recommended for publica-
tion by Associate Editor P. Dupont and Editor I. Walker upon evaluation of the
reviewers’ comments.
T. Imaida was with the Department of Mechanical Engineering, Graduate
School of Engineering, Kyoto University, Kyoto 606-8501 Japan. He is now
with Mitsubishi Heavy Industries, Ltd., Aichi 465–0042, Japan.
Y. Yokokohji and T. Yoshikawa are with the Department of Mechanical En-
gineering, Graduate School of Engineering, Kyoto University, Kyoto 606-8501
Japan (e-mail: yokokoji@mech.kyoto-u.ac.jp).
T. Doi was with the National Space Development Agency of Japan, Ibaraki
305-8505, Japan. He is now with Toshiba Corporation, Tokyo 105–8001, Japan.
M. Oda is with the National Space Development Agency of Japan, Ibaraki
305-8505, Japan.
Digital Object Identifier 10.1109/TRA.2004.825271
Lawn et al. [9] performed one-degree-of-freedom (DOF) tasks
such as pushing and positioning with time delay. They used a bi-
lateral control law based on the scattering theory and reported,
“The passivity-based laws were not tested for delays of 1 s since
their performance was very poor due to extremely low stiffness.”
Hirzinger et al. [2] mentioned that, “In ROTEX, the loop delays
varied from 5–7 s. Predictive computer graphics seems to be the
only way to overcome this problem.”
As Peñin et al. [15], [16] did, we also summarized previous
works on teleoperation with force feedback under certain com-
munication time delays in Table I [5]. All of them showed the
results of real experiments. These previous works can be di-
vided into two groups: 1) direct bilateral teleoperation without
any models of the remote site and 2) model-based teleoperation
with pseudo force feedback from a local model of the remote
environment. From the table, it seems that when the time delay
is longer than about 1 s, the model-based approach would be the
only solution. However, we have been doubtful about this “1-s
limitation” for the following reasons.
 Some of the observations came from the results using
a conventional bilateral controller, for which stability is
not guaranteed under the time-delay condition. Probably,
1 s would be the limitation to stabilize such an unstable
system by human operators.
 The bilateral control based on the scattering theory guar-
antees the stability of the system for any time delay. How-
ever, it loses its stiffness and tends to be sticky as the time
delay becomes large [13]. Again, 1–2 s would be the lim-
itation for an operator to maneuver such a system com-
fortably [9]. However, the scattering theory is not the only
solution to the time-delay problem, and some other types
of bilateral controller can also guarantee the stability.
Instead of exactly drawing the limitation line at 1 s, our claim
is, in a sense, quite natural as follows. The time-delay limitation
depends on the difficulty of the task. Even if the time delay be-
comes longer than 1 s, some tasks could be performed by direct
bilateral teleoperation. Actually, Ferrell [3] investigated the ef-
fect of time delays longer than 1 s in bilateral control. Although
the tasks he conducted were simple positioning with force feed-
back, he tested several time delays up to 3 s.
In this paper, the results of a ground–space teleoperation ex-
periment using a robot arm mounted on the Engineering Test
Satellite 7 (ETS-VII) are shown. The experiment was conducted
on November 22, 1999. Round-trip time for communication be-
tween the National Space Development Agency (NASDA) of
1042-296X/04$20.00 © 2004 IEEE