Preparation of Papers for the Journal
Engineering Letters

Abstract— The control and manipulation of a surgical
robot is an inherently challenging task especially when
estimating the actual force in real-time. The end-effector of
the surgical robot, where the surgical tools have been
attached, requires high-end precision. Most of the
commercial surgical robotic systems calculate the real-time
force by using traditional force sensors which
encounters hindrance like lack of expected response (advance
control), limited bandwidth and requirement of force for its
own operation. The paper introduces a Disturbance observer
(DOB) based Reaction Torque observer (RTOB) as the sensor
for the real-time force sensing in biomedical applications,
with a focus on surgical robots. Besides, the force sensation
DOB would also eliminate any unwanted disturbances. In the
experiment, RTOB was implemented with the real-time
surgical robot for the bandwidth of 0-80 Hz. The
experimental results corroborate the importance and
usefulness of RTOB for novice researchers.
Index Terms—About four key words or phrases in
alphabetical order, separated by commas, for example,
visual-servoing, tracking, biomimetic, redundancy, degrees-
of-freedom
I. INTRODUCTION
Robot-assisted surgery will continue to accept and
incorporate developments in a variety of main areas in the
foreseeable future .It allows physicians to perform many
forms of complex procedures with greater precision,
versatility and control than traditional techniques. As an
advancement, surgical robot for minimally invasive
surgery(MIS) has been widely researched and
developed .During MIS procedures, as the surgeon is
separated from patient, he experience a lack of direct
palpation ,tactile sensation , poor visibility, poor hand-eye
coordination, inadequate depth details, no proper force
feedback from touching the organs and movement
limitations[1]. Most of the prevalent MIS systems
incorporates bilateral control systems (master and slave
robots). In the conventional bilateral control, much
research has paid attention to develop control architectures,
 lends a unique attribute of being sensorless without
Manuscript received June XX, 20XX; revised July XX, 20XX. (Write
the date on which you submitted your paper for review.) This work was
supported in part by the U.S. Department of Commerce under Grant
BS123456 (sponsor and financial support acknowledgment goes here).
Paper titles should be written in uppercase and lowercase letters, not all
uppercase. Avoid writing long formulas with subscripts in the title; short
formulas that identify the elements are fine (e.g., "Nd–Fe–B"). Do not write
“(Invited)” in the title. Full names of authors are preferred in the author
field, but are not required. Put a space between authors’ initials.
F. A. Author is with the National Institute of Standards and Technology,
Boulder, CO 80305 USA (corresponding author to provide phone: 303-
555-5555; fax: 303-555-5555; e-mail: author@ boulder.nist.gov).
S. B. Author was with Rice University, Houston, TX 77005 USA. He is
now with the Department of Physics, Colorado State University, Fort
Collins, CO 80523 USA (e-mail: author@lamar.colostate.edu).
T. C. Author is with the Electrical Engineering Department, University
of Colorado, Boulder, CO 80309 USA, on leave from the National
Research Institute for Metals, Tsukuba, Japan (e-mail: author@nrim.go.jp).
and implemented force/torque sensor to detect external
force However, it is inappropriate to attach any electric or
magnetic sensor for measuring force on surgical robot
instrument because the surgical robot instrument should be
manufactured by small size and overcome safety problem
caused by using electric or magnetic sensors in human
body [2].
The traditional force/torque sensors used in surgical
robots are supposed to be positioned at different places
which include actuator, gripper surface, or in an
appropriate site and their range of force sensation will be
restricted to the places where they are installed. Some
inherited properties of these force sensors are responsible
for causing errors in the output, e.g. the physical force
sensor tends to add inertia or mass to the system, which
will not be considered by the sensor during the estimation
of reaction force. Additionally, they have narrow
bandwidth and hence the measured force is significantly
dull. Therefore, it is clear that the force sensors are not
suitable with applications that demand high precisions like
robotic surgery. In order to keep up with the current
requirements, the design of sensor systems is expected to
include innovative approaches and solutions benefiting
from recent scientific and technological developments.
Consequently, the advancement of sensor technology is
focused on the permanent technological progress in these
areas. In particular, a major upturn in these fields has been
observed in recent years, involving a great potential for
entirely new sensor and sensor systems approaches.
Taking into considerations the impediments imposed by
the application requirements, Murakami et al proposed a
force control method known as reaction torque/force
observer (RTOB) [3]. The method of force control is based
on disturbance observer (DOB), which is a common
method to estimate and compensate the disturbance force.
A RTOB/RFOB system excludes system uncertainties such
as friction from the disturbance input to estimate the
reaction torque/force. This reaction torque/force forms a
basis for feedback in a force control system without the use
of any force sensors. Considering the implementation of
robots in medical applications, which demands an
appropriate force control method, the DOB-based RTOB
compromising the performance.
First of all, as showed by [4] [5], the sensorless force
sensing system offers a wider bandwidth as compared to
the conventional method. On the one hand, wider
bandwidth allows for proper response, and on the other, the
system stability is improved significantly. Secondly, the
RTOB system demonstrates substantial consistency with
the variations in mechanical structures [5], as it entirely
excludes the mechanical constraint of a force sensor. In
addition, [5] has also highlighted some shortcomings with
implementation of a force sensor such as high noise,
inaccuracy due to natural frequency and a change in system
structure due to the force sensor(which typically is a strain Equation (4) expressed as
gauge) being soft. Moreover, the system also comprises of (5)
an integrator for the subtraction of the steady state errors
[7]. Considering numerous such studies, it is pertinent that
the RTOB system provides robust solution in the (6)
applications which demand a meticulous motion control.

II. MATERIALS AND METHOD

TL
- l
A. DC Motor Model + 1  1 
Robotics applications has been dominated by the use of a Tm
DC Motor as an actuator. Same applies regarding the m Js s
scenario of a surgical robot as well. Therefore, the proper
modeling of a DC motor would be a convenient point to Figure 3. Block diagram of a servo motor
initiate. General diagram of a DC motor has been
demonstrated in Figure 2. Where, is a sum of inertia torque, external torque and
torque generated from coulomb and viscous friction.

(7)

Where;
: Inertia torque and gravity effect
: External torque
: Static friction
Figure 1. DC Motor equivalent circuit diagram
: Viscous friction

The reference current is the input for the motor and

Applying Kirchoff’s Voltage Law,
PWM voltage pulses are responsible for the control of the
(1) motor. For correct functionality, must be compensated,
And, (2) as the motor torque cannot be directly controlled using

(3) . As the torque disturbs the output torque, it is also

termed as ‘Disturbance Torque’. Hence, ‘Disturbance
The torque of motor expressed as, Observer’ is introduced to measure this disturbance torque
and compensate it [13].
(4)
Where, B. Disturbance Observer modelling
= Inertia of the motor
Equation (4) consists of a couple of parameters motor
= Viscous friction coefficient inertia and torque constant, which can be changed because
= Load torque of several properties. Similarly, due to the mechanical
configuration of the motion system, the inertia might also
= Static friction torque change. In the case of the torque coefficient, it changes
= Angular speed with respect to the rotor position of the electric motor,
which causes the irregularity in the distribution of the
= Torque constant magnetic flux throughout the surface of the rotor [11, 13].
Tl  T f
I aref

Va  1  1  (8)
kt Tm B  Js
R  Ls

(9)
Back Emf
Where;
ke
Nominal inertia
Figure 2. Block diagram of DC Motor
Inertia variation
Nominal torque coefficient
 Variation of torque coefficient implementation of the disturbance observer in the DC
motor observed in the following figure 6.
By substituting equations (8) and (9) into equation (4),
equation (10) obtained:
I aref   
K tn g dis J n
(10)

By rearranging,
g dis
s  g dis
(11)

Where;

Disturbance Torque g dis J n

Therefore, Tˆdis

Figure 5. Block diagram of the disturbance observer

(12)

Apparently, the disturbance torque comprises all the Tdis

internal torques, external torques, friction torques and

torques due to parameter variations and all of them I aref  1  1 
Kt
compensated so that the accurate output torque obtained. Js s

From equation (11) can be calculated as follows.
  DC Motor
K tn g dis J n
(13)
Disturbance
I aref    1 g dis Observer
K tn Jns s  g dis
K tn

Tdis 

g dis J n
Differentiator
g dis
Tˆdis
s  g dis
Figure 6. Disturbance observer with the DC motor
T̂dis
C. Reaction torque observer modeling
Figure 4. Estimated disturbance through a low pass filter
The significance of the disturbance detected in practical
Using the known values at the right-hand side, the
application is observed as disturbance compensation as
unknown value of can be determined using the known well as in the reaction torque estimation [1, 16]. Without
values at right-hand side in the above equation. Using the using a torque sensor, the disturbance observer can identify
velocity response and the torque current , the the internal disturbance of the system and also estimate the
reaction torque as shown in figure 7. It is well accounted
estimated disturbance torque is obtained. As shown in that the transmission of force sensation within bilateral
equation (14), the estimation is done through a first-order control is dictated by their action and reaction relationship.
low-pass filter, where represents the cut-off frequency In this case, by increasing the cut-off frequency, prompt
calculation and estimation of the reaction torque are
of the low-pass filter. Low-pass filter to filter the noises
commenced by the disturbance observer. The reaction
generating from the differentiation [15].
torque observer is considered as a modification of the
disturbance observer.
(14) . (15)

The equations represented in the form of a block diagram.
The equation (12) and (13) demonstrated as a block
diagram as shown in Figure 5. This structure comprises of
a differentiator, which will escalate the noise. Considering
that, the same diagram restructured as follows. The
The external torque acting on the slave manipulator
considered as the Reaction torque and it calculated as:
(16)
Since at the moment of contact, bandwidth of 0-80 Hz, which is sufficient for responding to
the control commands of surgeons.
(17)
Where: Frequency Response curve of the system
Reaction Torque

(18)

The block diagram of the reaction torque observer shown

in figure 8.

Tdis


I aref  1  1 
Kt
Js s

The two main units of this experimental setup are master

 
K tn g dis J n
DC Motor and slave surgical manipulators. The master side was
 Tdis
operated by the surgeon. For an ideal master-slave system
 the applied force and the position of the master should be
Tint  ( f  D )  J  K t I aref Reaction Torque
Observer replicated at the slave system. In experiment, the results of
master-slave control are shown separately in terms of
position and force applied. The master-slave position and
g dis
force response at different joints are demonstrated in
s  g dis
figures xx, xx, xx, xx & figures xx, xx, xx, xx.


g dis J n

Tˆreact

Figure 7. Reaction Torque Observer with the DC motor

The disturbance observer offers numerous advantages

including observation of disturbance torque and
compensation. While a regular force sensor consists of
narrow bandwidth, the disturbance observer offers Figure:
relatively wider bandwidth [15]. The force sensor
consisting of narrow bandwidth cannot offer substantial
force sensation and is just able to detect wherever it is
implemented. Due to being to operate under a smaller
sampling time, which increases the observer gain as much
as possible, the disturbance observer can provide a better
alternative with the wider bandwidth. As such, the
disturbance observer is relatively more appropriate for
transmission of force sensation in comparison to the force
sensors.

III. RESULTS AND DISCUSSION

Figure:

In order to test the frequency response range of the

master and slave system Fast Fourier Transform Test has
been conducted on the real-time data obtained from the
experiment conducted by the surgeon. In the experiment,
RTOB was implemented with the real-time surgical robot.
FFT (Fast Fourier Transform) analysis was carried out with
the force information collected from the surgical robots
during the period that the master and slave made mirror
symmetric movements and it was found that results
obtained were the same for the data detected on the master
and slave surgical robots. This result shows RTOB can be
implemented with the real-time surgical robot for the
Figure:

Figure:

There is a force lock system in order to control the slave at

commence of operation. The activated force controller
maintains the slave position until the master applies the
releasing force. The position of the master during the force
Figure: locked period on the slave is controlled by a spring
controller which is appended onto the master. Apart from
the force lock mode, master and slave operate under
bilateral control. The results shown in fig depicted a force
error of 5.26% and a position error of 0.8% estimated using
standard linear regression method. For sudden changes we
can observe a small lag (0.2 seconds) for the response of
slave system. In the case of positive direction of rotation it
has a maximum error of 0.005 rad/sec. For the case of
negative direction of rotation it has a maximum error of
0.002 rad/sec.

IV. CONCLUSION
A conclusion section is not compulsory. Although a
Figure: conclusion may review the main points of the paper, do not
replicate the abstract as the conclusion. A conclusion might
elaborate on the importance of the work or suggest
applications and extensions.

ACKNOWLEDGMENT

REFERENCES
[1].Branesh M. Pillai, Chumpon Wilasrusmee, Jackrit
Suthakorn,"Observe based dynamic control model or
bilaterally controlled MU-Lapa robot: Surgical tool force
limiting", International Journal of Electrical and Computer
Engineering (IJECE), 2020.
[2]. Sung Min Yoon; Won Jae Kim; Min Cheol Lee; Woo
Hyeok Choi, "Bilateral Control for Haptic Laparoscopic
Surgery Robot", IEEE ISR, 2013
[3] T. Murakami, F. Yu and K. Ohnishi, "Torque
Sensorless Control in Multidegree-of-Freedom
Manipulator," IEEE TRANSACTION ON INDUSTRIAL
ELECTRONICS, pp. 259-265, 1993.
[4] S. Katsura, Y. Matsumoto and K. Ohnishi, "Analysis
and experimental validation of force bandwidth for force
control," IEEE TRANSACTIONS ON INDUSTRIAL
ELECTRONICS, vol. 53, no. 3, pp. 922-928, 2006.
[5] Y. M. a. K. O. S. Katsura, "Modeling of Force
Sensing and Validation of Disturbance Observer for Force
Control," IEEE Transactions on Industrial Electronics, vol.
54, no. 1, pp. 530-538, 2007.
[6] L. Y. F. and C. X. B., "On the Dynamic Behavior of a
Force/Torque Sensor for Robots," IEEE TRANSACTIONS
ON INSTRUMENTATION AND MEASUREMENT, vol.
47, no. 1, pp. 304-308, 1998.
[7] S. E. and O. K., "Robust force control via disturbance
observer," IECON 2013 - 39th Annual Conference of the
IEEE Industrial Electronics Society, pp. 6551-6556, 2013.

First A. Author (M’76–SM’81–F’87) and the other authors may include

biographies at the end of regular papers. Biographies are often not included
in conference-related papers. This author became a Member (M) of IAENG
in 1976, a Senior Member (SM) in 1981, and a Fellow (F) in 1987. The
first paragraph may contain a place and/or date of birth (list place, then
date). Next, the author’s educational background is listed. The degrees
should be listed with type of degree in what field, which institution, city,
state, and country, and year degree was earned. The author’s major field of
study should be lower-cased.
The second paragraph uses the pronoun of the person (he or she) and
not the author’s last name. It lists military and work experience, including
summer and fellowship jobs. Job titles are capitalized. The current job must
have a location; previous positions may be listed without one. Information
concerning previous publications may be included. Try not to list more
than three books or published articles. The format for listing publishers of a
book within the biography is: title of book (city, state: publisher name,
year) similar to a reference. Current and previous research interests end the
paragraph.
The third paragraph begins with the author’s title and last name (e.g., Dr.
Smith, Prof. Jones, Mr. Kajor, Ms. Hunter). List any memberships in
professional societies other than the IAENG. Finally, list any awards and
work for IAENG committees and publications. If a photograph is provided,
the biography will be indented around it. The photograph is placed at the
top left of the biography. Personal hobbies will be deleted from the
biography.