Control of Nonlinear Bilateral Teleoperation Systems Subject to
Disturbances
Alireza Mohammadi, Mahdi Tavakoli and Horacio J. Marquez
Abstract— Teleoperation systems, consisting of a pair of
master and slave robots are subject to different types of
disturbances such as joint frictions, varying contact points, un-
modeled dynamics and unknown payloads. Such disturbances,
when unaccounted for, cause poor teleoperation transparency
and even instability. This paper presents a novel nonlinear
bilateral control scheme, based on the concept of disturbance
observer based control, to counter these disturbances and their
negative effects on the teleoperation systems. The proposed
disturbance observer based bilateral control law is able to
acheive global asymptotic force tracking, and global exponential
position and disturbance tracking in the presence of various
disturbances. The minimum exponential convergence rate of
the position and the disturbance tracking errors can be tuned
by the controller parameters. Simulations are presented to show
the effectiveness of the proposed control scheme.
I. INTRODUCTION
Every teleoperation system consists of a master robot and
a slave robot. The master interacts with a human operator
and the slave interacts with a remote environment. If force
feedback from the slave side to the master side is provided,
the system is called a bilateral teleoperation system to dis-
tinguish it from a unilateral teleoperation system. A bilateral
teleoperation system is said to be transparent if the slave
robot precisely follows the position of the master robot and
the master robot faithfully displays the slave-environment
contact force to the human operator.
The most successful control scheme in achieving a fully
transparent teleoperation system is the 4-channel architecture
[1], [2], which is mostly suitable for teleoperation systems
with fixed linear models. Physical robots, however, are
nonlinear systems subject to various disturbances, such as
unknown dynamics, joint frictions, unknown payloads, etc.
[3]. Such disturbances, when unaccounted for, cause poor
teleoperation transparency and even instability. One way
to suppress these disturbances, is to employ disturbance
observers [4]. Disturbance observer based control has been
used in applications such as independent robot joint control
[5], and robot joint friction estimation and compensation [6],
[7].
A considerable part of the existing literature on disturbance
observer design for robotic applications uses linearized mod-
els or linear system techniques [8], [9]. In order to overcome
This work was supported by the Natural Sciences and Engineering
Research Council of Canada (NSERC).
A. Mohammadi was with the Department of Electrical & Computer
Engineering, University of Alberta, Edmonton, AB T6G 2V4 Canada. He is
now with the Edward S. Rogers Sr. Department of Electrical & Computer
Engineering, University of Toronto, Toronto, ON M5S 3G4 Canada. Email:
alireza.mohammadi@mail.utoronto.ca
M. Tavakoli is with the Department of Electrical & Computer Engi-
neering, University of Alberta, Edmonton, AB T6G 2V4 Canada. Email:
tavakoli@ece.ualberta.ca
H. J. Marquez is with the Department of Electrical & Computer Engi-
neering, University of Alberta, Edmonton, AB T6G 2V4 Canada. Email:
marquez@ece.ualberta.ca
the linear disturbance observer limitations for the highly
nonlinear and coupled dynamics of robotic manipulators,
Chen et al. proposed a nonlinear disturbance observer for
nonlinear robotic manipulators and designed it such that
no acceleration measurement was needed [10]. However,
the closed-loop stability of the overall system including the
disturbance observer and the controller was not investigated.
The investigation of the stability and performance of a
master-slave teleoperation system under disturbance observer
based control is even more challenging and not studied
either. While a distrubance observer based controller was
designed for bilateral teleoperation systems in [11], the
master and the slave robots were considered to be linear
with only one DOF. This serves as the motivation to look
for a disturbance observer based control law for nonlinear
and n-DOF teleoperation systems.
This paper
1
addresses the problem of disturbance observer
based control of nonlinear teleoperation systems. A dis-
turbance observer based control law will be proposed and
incorporated into the framework of the 4-channel teleopera-
tion architecture. Under the proposed control law, full trans-
parency and exponential disturbance and position tracking
are achieved.
The organization of this paper is as follows. Section II
introduces the nonlinear model of the teleoperation systems
and 4-channel bilateral control architecture. Section III pro-
poses a novel disturbance observer based controller for non-
linear teleoperation systems subject to various disturbances.
The teleoperation system transparency is also addressed in
this section. Finally, simulations in section IV show the
efficiency of the proposed control scheme as compared with
the case where no disturbance observer is employed.
II. NONLINEAR MODEL OF A TELEOPERATION
SYSTEM
The dynamical models investigated in this paper and our
proposed control law in section III will be in the Cartesian
space. This enables us to make a teleoperation system
transparent without requiring the master and the slave robots
to have similar kinematics and dynamics.
A. Model of a teleoperation system in the Cartesian space
The dynamic equations describing the motions of the end-
effectors of n-DOF master and slave robots, which interact
with the human operator and the remote environment, in the
presence of disturbances can be written as
M
xm
(q
m
) ¨ x
m
+ N
xm
(q
m
, ˙ q
m
) = f
m
+ f
h
+ d
m
(1)
M
xs
(q
s
) ¨ x
s
+ N
xs
(q
s
, ˙ q
s
) = f
s
- f
e
+ d
s
(2)
1
A more complete version of this paper can be found in [12].
2011 50th IEEE Conference on Decision and Control and
European Control Conference (CDC-ECC)
Orlando, FL, USA, December 12-15, 2011
978-1-61284-799-3/11/$26.00 ©2011 IEEE 1765