Journal of Robotics and Control (JRC)

Volume 5, Issue 1, 2024

ISSN: 2715-5072, DOI: 10.18196/jrc.v5i1.20524 132

Design and Implementation of Fuzzy Logic

for Obstacle Avoidance
in Differential Drive Mobile Robot
Riky Dwi Puriyanto 1* , Ahmad Kamal Mustofa2
1,2
Dept. of Electrical Engineering, Universitas Ahmad Dahlan, Indonesia
Email: 1 [email protected], 2 [email protected]
*Corresponding Author

Abstract—Autonomous mobile robots based on wheel drive are can be used in all terrains. Mobile robots also play an important
widely used in various applications. The differential drive mobile role in education as they provide a flexible platform to explore
robot (DDMR) is one type with wheel drive. DDMR uses one and teach various topics such as mechanics, electronics, and
actuator to move each wheel on the mobile robot. Autonomous
capabilities are needed to avoid obstacles around the DDMR. This software. Mobile robots are widely used because they are the
paper presents implementing a fuzzy logic algorithm for obstacle smallest and have relatively lower investment costs than flying
avoidance at a low cost (DDMR). The fuzzy logic algorithm and humanoid robots. Apart from that, the way of movement
input is obtained from three ultrasonic sensors installed in front used by mobile robots is also understood by humans. This
of the DDMR with an angle difference between the sensors of makes it easier to develop mobile robots that use wheels as
450 . Distance information from the ultrasonic sensors is used to
regulate the speed of the right and left motors of the DDMR. propulsion in various applications.
Based on the test results, the Mamdani inference system using One of the developments in mobile robots in the industry
the fuzzy logic algorithm was successfully implemented as an 4.0 era is autonomous mobile robots (AMR). Advances in
obstacle avoidance algorithm. The speed values of the right and
left DDMR wheels produce values according to the rules created in
sensors, data processing, and artificial intelligence have enabled
the Mamdani inference system. DDMR managed to pass through AMRs to become more sophisticated, adaptive, and flexible in
a tunnel-shaped environment and reach its goal without hitting carrying out assigned tasks. One of the capabilities AMR needs
any obstacles around it. The average speed produced by DDMR to prevent damage from collision hazards is obstacle avoidance
in reaching the goal is 4.91 cm/s. [13]. Obstacle avoidance capabilities to detect, identify, and
avoid obstacles efficiently [14]. This is the basis for the success
Keywords—DDMR; obstacle avoidance; mamdani; fuzzy logic; of autonomous vehicles and robots operating in dynamic envi-
mobile robot.
ronments [15]. Robots that can avoid obstacles will be able to
reduce the risk of collisions associated with financial investment
I. I NTRODUCTION in robot development.
Mobile robots are one of the successful achievements in One of the algorithms used in obstacle avoidance is fuzzy
robotics and provide a path to a new era in automation technol- logic (FL) [16]–[20]. Fuzzy algorithms utilize an approach like
ogy. Advances in artificial intelligence, sensory, and movement how humans make uncertain decisions [20]. Fuzzy logic allows
technology mean mobile robots can move autonomously with robots or autonomous systems to make decisions based on
increasingly high levels of accuracy. Mobile robots are widely understanding the ”truth” or ”untruth” of a condition [18]. This
implemented in various fields such as industry [1]–[3], health algorithm allows the system to listen and respond to sensory
[4]–[6], military [7]–[9] and education [10]–[12]. In the indus- information in a given condition [16]. The use of fuzzy logic
trial sector, mobile robots can be used to improve automation in obstacle avoidance allows robots to make decisions based on
systems and perform repetitive work precisely and at lower information, such as distance and speed [17] and can respond
costs. The health sector also uses mobile robot technology in to the information obtained with an appropriate level of caution
services that help patients. The use of mobile robot services is [19]. A control system that uses fuzzy logic as a controller can
very beneficial during the COVID-19 outbreak because it not be built more simply and flexibly to handle the system without
only prevents the spread of infection and reduces human error having to build a mathematical model. This is the advantage of
but also allows health staff to reduce direct contact. Mobile fuzzy logic, so it is widely used in various applications. One
robots can be used to improve navigation techniques so that they application that can be solved using the fuzzy logic algorithm
is obstacle avoidance on mobile robots.

Journal Web site: http://journal.umy.ac.id/index.php/jrc Journal Email: [email protected]

Journal of Robotics and Control (JRC) ISSN: 2715-5072 133

One important part of fuzzy logic is the fuzzy inference mechanical or electronic entities that can perform specific tasks
system (FIS). One type of fuzzy inference system that is automatically, with or without human intervention [29]. In other
considered to have advantages in a more interpretable rule words, robots are automatic machines that can replace human
base is Mamdani [16], [21], [22]. Mamdani FIS is one of work, even though they do not have a human-like appearance
the most used approaches in applying fuzzy logic in decision- or do not carry out tasks like humans do [30]. The use of
making [16]. Mamdani is also considered more popular than robots has expanded in various sectors of human daily activities.
Sugeno and Tsukamoto. One of the main characteristics of the These sectors include manufacturing, transportation, regional
Mamdani fuzzy inference system is its ability to develop rules exploration, medical fields, military needs, and laboratory ex-
that can be easily interpreted by humans [21]. This allows periments. Robots are generally divided into mobile robots
domain experts, such as engineers or experts in a particular and fixed robots [31]. The difference between the two lies
field, to creatively construct rules based on their knowledge in the ability of movement and destruction in carrying out it
of the system organized in a fuzzy environment. Apart from [32]. Robots of the fixed robot type work in a predetermined
that, the Mamdani fuzzy inference system is also flexible in and fixed environment [33]. Meanwhile, mobile robots operate
handling devices in data input [22]. In the Mamdani method, in a constantly changing environment [34]. Fixed robots are
all input and output variables have a membership function that more frequently used in industry for repetitive tasks, and the
accommodates all members of the fuzzy set [23]. location of use has been previously identified, such as robot
Mamdani’s fuzzy inference system is a type of fuzzy infer- manipulators that inspect materials moving on conveyors [34].
ence system that is very suitable to be applied in the context of Meanwhile, mobile robots must carry out more complex and
obstacle avoidance on mobile robots. The information required dynamic tasks quickly because they operate in unpredictable
by the obstacle avoidance algorithm is generally obtained from environments [33].
sensors installed on the robot body. These sensors will act as Fixed robots operate in a fixed position or with minimal
the robot’s eyes and ears so that it can provide insight and movement [35]. Fixed robots are usually designed to perform
understanding of the surrounding environment. Several sensors specific tasks in a relatively static environment. One of the
used for obstacle detection in mobile robots include ultrasonic most significant advantages of fixed robots is the robot’s
[24]–[26], lidar [27], [28], and others. The ultrasonic sensor will ability to work tirelessly and consistently, thereby increasing
work by utilizing ultrasonic sound waves emitted and reflected productivity. Fixed robots can handle monotonous and repetitive
by surrounding objects [24]. This can be used to measure the tasks, allowing human workers to focus on more complex and
distance between a robot and an object. Ultrasonic sensors are creative aspects of work [36]. Additionally, fixed robots are
very useful in detecting obstacles at short to medium distances adaptable as they can be reprogrammed to perform different
[25] and in complex environments [26]. In the implementation tasks, which makes them cost-efficient for businesses with
of mobile robots, ultrasonic sensors are widely used because growing production needs [37].
they have the advantage of low cost. Meanwhile, lidar uses Even though they have many advantages, fixed robots also
laser technology which can create a three-dimensional map of have their challenges. One significant problem is the high initial
the surrounding environment [27]. This gives the robot a higher costs for acquisition and installation, which can be a barrier for
level of accuracy in detecting objects at long distances [28]. small businesses [38]. Additionally, programming a fixed robot
The information collected by these sensors will be used by the to perform a new task can be time-consuming and requires
obstacle avoidance algorithm to make decisions about how the skilled technicians [39]. Fixed robots can be found in various
robot should move. applications in the industrial world, such as manufacturing,
The contribution of this research is to design a fuzzy logic- automotive, pharmaceuticals, and others.
based obstacle avoidance algorithm using three ultrasonic sen- Meanwhile, mobile robots, which are also known as au-
sors to control the speed of the mobile robot. This research also tonomous robots or mobile autonomous robots, are a type of
uses a low-cost mobile robot platform to implement a real-time robot that can move or move places [40]. In contrast to fixed
obstacle avoidance algorithm. robots, which are stationary, mobile robots are equipped with
This article is organized as follows: Section 2 explains the mobility mechanisms that allow them to move and navigate the
robot design used in this research. Section 3 presents the environment independently. This autonomy is made possible
Fuzzy Logic method used to avoid obstacles in the testing through sensors, cameras, and advanced algorithms that enable
environment. Section 4 presents the results of the experiment. robots to sense their surroundings, make decisions, and move
The conclusion of this study is presented in Section 5. across complex terrain. Mobile robots often have wheels, legs,
wings, and propellers designed for specific applications and
II. M OBILE ROBOT environments. Mobile robots, such as autonomous vehicles,
drones, and exploration, are often used in various applications
Advances in robot technology have become an important
that require mobility and adaptation to environmental changes
factor in the world of technology and industry. Robots are

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 134

[40]. The flexibility and adaptability of mobile robots make locations, maintaining operator safety. Mobile robots are more
them a dynamic and transformative technology with the poten- flexible in various applications [49]. Mobile robots can be
tial to revolutionize various sectors. customized and configured to perform various tasks, while fixed
Mobile robots can be divided according to where they move robots tend to be designed for specific tasks. Mobile robots
into three main categories: aerial (air), underwater (underwater), have higher manoeuvrability in carrying out tasks that require
and terrestrial (land) [41]–[43]. Each category has specific movement around objects, obstacles, or complex situations and
characteristics and is used in various applications based on the can adapt to changing conditions more efficiently compared to
environment in which it moves. Aerial robots are a type of fixed robots than fixed robots [45].
mobile robot designed to move and operate in the air or an One type of mobile robot is a differential drive mobile robot,
open environment [41]. A drone or unmanned aerial vehicle which uses a driving method based on differences in the speed
(UAV) is the most common example. Drones are used in of wheels placed on opposite sides [50]. The differential drive
various applications, such as aerial mapping, surveying, aerial mechanism allows this robot to move independently by control-
photography, surveillance, and entertainment. The advantage of ling the right and left wheels’ speed [51]. This creates flexible
aerial robots is the robot’s ability to reach locations that are robot manoeuvres and allows the robot to easily perform various
difficult to reach or potentially dangerous for humans and can movements such as turning, turning, or going back and forth.
provide a unique perspective from the air [41]. Differential drive mobile robots are often equipped with wheels
Furthermore, underwater robots, such as Remotely Operated that can be rotated independently so that the robot can precisely
Vehicles (ROV) or Autonomous Underwater Vehicles (AUV), regulate the orientation and direction of the robot’s movement
move and operate below the water surface, such as in the sea, [52]. Therefore, these robots are suitable for navigation in
lake, or river [42]. Underwater robots are used for deep sea cramped or complicated environments, such as warehouses,
exploration, underwater environmental monitoring, marine re- laboratories, or even on exploration missions in unpredictable
search, seafloor mapping, and other tasks that require operations environments.
below the water surface [42]. Also, underwater robots are often The advantages of differential drive mobile robots are sim-
used in extreme deep-sea exploration and can reach depths plicity in design and construction and good navigation capabil-
that human divers cannot reach. Furthermore, terrestrial robots ities in confined environments [53]. This robot is often used or
move on land surfaces, such as on land, roads or the earth’s applied in automatic sweeping robots, goods delivery devices,
surface [43]. This type of robot includes various varieties, such or even robotics contests. However, differential drive mobile
as autonomous cars, rover robots, and service robots, used in robots also have several limitations, especially in overcoming
transportation, monitoring, agriculture, and facility maintenance rough or uneven terrain [54]. To overcome this problem, some
applications. The advantages of terrestrial robots are the ability differential drive mobile robots are equipped with distance-
to operate in various terrestrial environments and existing measuring sensors or additional navigation systems to increase
infrastructure, as well as flexibility in dealing with various the robot’s ability to explore more complex environments [54].
challenges that may be encountered on land [44]. The low-cost differential drive mobile robot (DDMR) design
Mobile robots have several advantages compared to fixed used can be seen in Fig. 1. The low-cost DDMR design was
robots, mainly depending on the task type to perform and the created by considering the manufacturing costs and function of
work environment [45]. Mobile robots such as autonomous the DDMR. DDMR consists of three ultrasonic sensors facing
cars and drones can move places compared to fixed robots forward with an angle difference between the sensors of 450 .
with limited movements [45]. This allows the mobile robot to The three ultrasonic sensors will be used to detect the distance
reach different locations quickly and flexibly, according to task of obstacles in front of DDMR. DDMR uses two DC motors as
requirements. Mobile robots are more adaptive to environmental actuators. The wheel revolutions are measured with two external
changes than fixed robots [46]. Mobile robots can navigate and encoders equipped with encoder disks.
interact with changing environments in the field, outdoors, and In this research, low-cost DDMR is given obstacle avoidance
dynamic external locations. capabilities to avoid obstacles while moving toward the target.
Meanwhile, fixed robots are usually designed to operate in Cheap sensors are expected to function optimally in detecting
a stable or well-defined environment. Mobile robots are very obstacles. The relationship between the input, microcontroller,
suitable for monitoring and exploration, especially in environ- and DDMR actuator can be seen in Fig. 2. Three ultrasonic
ments that humans cannot reach, such as using drones in forest sensors are connected to a microcontroller to obtain the distance
mapping, underwater probe robots for deep sea exploration, value between the robot and the obstacle. DC motor rotational
and space rovers for exploring planets [47]. Mobile robots can speed data is obtained from two encoders installed on the right
be used for rescue tasks in dangerous environments, such as and left sides of the robot. The microcontroller carries out speed
search and rescue in natural disasters, fires, or other dangerous regulation to the DC motor via the motor driver.
zones [48]. Mobile robots can be operated away from risk

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 135

of truth of a statement. In fuzzy logic, mathematical and logical

operations such as conjunction (AND), disjunction (OR), and
negation (NOT) are redefined for use in the context of fuzzy
sets [59]. Apart from that, there is the term fuzzy inference,
which is a decision-making process based on fuzzy rules which
are explained in the form of ”fuzzy rules” or ”IF-THEN rules”
[60]. Each direction describes the relationship between input
variables and output variables in the form of a fuzzy set. The
fuzzy inference system then combines these rules to produce a
final decision [61].
(a) 3D design of DDMR
Fuzzy logic deals with uncertainty and subjectivity in mod-
elling and analysis, which are often difficult or impossible to
represent in traditional ways. The main concept in fuzzy logic
is the use of fuzzy sets, which replace conventional sets, which
are strict [62]. Fuzzy logic is often used in the field of robotics
[63]. Applying fuzzy logic concepts in robotics allows robots to
overcome uncertainty and complexity in decision-making [64].
Fuzzy logic also replaces conventional approaches based on
Boolean logic, which only recognizes true or false values [65].
The fuzzy logic approach allows the representation of various
levels of truth, from true to false. This causes robots to behave
more humanely and adaptively in various situations [66].
The fuzzy inference system is an important part of fuzzy
logic in developing mobile robots [67]. Robots are often faced
(b) Design of DDMR with complex and uncertain situations, such as environmental
Fig. 1. Design of low-cost differential drive mobile robot changes, sensor input variations, and decision-making uncer-
tainty [68]. A fuzzy inference system (FIS) is a key component
in developing robots that overcome uncertainty and complexity
in decision-making and can change environments [69]. Using a
fuzzy inference system in a mobile robot allows a more adaptive
and responsive decision-making process based on sensor data,
which is often vague or unclear [70]. Mobile robots, often
used in various applications such as exploration, logistics, and
customer service, need this adaptability.
One important application is using a fuzzy inference system
Fig. 2. Block diagram of differential drive mobile robot in mobile robot navigation [71]. Fuzzy inference systems are
used to process sensor data such as distance, direction, and
image data from cameras to guide robot movement and avoid
III. M ETHODS obstacles [72]. Robots that use a fuzzy inference system can
make decisions based on how far the robot approaches an object
Fuzzy logic is a method in computer science and mathematics
or obstacle so that the robot can move more safely and avoid
that allows the processing of information that is not clear or
collisions [73]. This allows robots to deal with diverse situa-
exact [55]. This concept was first proposed by Lotfi Zadeh in
tions, such as passing through narrow passageways, interacting
1965 and has been applied in various fields, such as automatic
with users, or moving in unpredictable environments. It can also
control decision-making and artificial intelligence [56]. Fuzzy
be used in mobile robots that must make decisions about speed
logic is based on the idea that many variables and concepts in
and direction to reach a destination without exact knowledge
the real world cannot be expressed strictly or in binary [57].
of road conditions [74].
Instead, these elements have degrees of membership in a fuzzy
set, which allows measuring truth on a continuum between true Fuzzy inference systems are also used in user-based decision-
and false. Fuzzy logic uses membership functions to describe making [75]. Robots equipped with fuzzy inference systems
the extent to which an element is included in a fuzzy set [58]. can understand human instructions or preferences better than
Membership functions can be triangular, trapezoidal, or others robots that only recognize ”right” or ”wrong” commands [76].
that suit the problem context. This function determines the level This allows for more natural and efficient interactions between

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 136

humans and robots. Mobile robots also use fuzzy logic to


overcome complex situations on the road. In complex track sit- 0 if xi < 10

uations, fuzzy logic allows vehicles to make decisions based on 
xi −8
10 ≤ xi ≤ 15


 7
 if
finer levels of truth, thereby avoiding accidents and optimizing
performance in various conditions [77]. µmedium (xi ) = 1 if 15 < xi ≤ 30 (2)
37−xi

if 30 < xi ≤ 37


Using a fuzzy inference system in a mobile robot is also very 

 7
0 if xi > 37
important in an automated delivery system because it allows
it to make adaptive and responsive decisions in managing the
delivery of goods [78]. Fuzzy inference systems help robots 
make decisions about optimal routes, avoid obstacles, and adjust 1
 if xi < 33
xi −33
speed based on changing situations, enabling more efficient µf ar (xi ) = if 33 ≤ xi ≤ 40 (3)
 7
delivery of goods [79]. In addition, mobile robots used in 
0 if x > 40
environmental exploration, such as underwater probe robots
or rovers on other planets, rely heavily on fuzzy inference
systems to deal with uncertainty in decision-making]. Fuzzy
inference systems enable robots to move safely in dangerous
or unpredictable environments, respond to changing conditions,
and achieve exploration goals [80].
In this research, a fuzzy inference system design based on
the Mamdani method will be designed, which consists of three
inputs and two outputs. The input consists of three ultrasonic
sensors, which are defined as the left sensor, middle sensor,
and suitable sensor. The output used in this research is two DC
(a) Left sensor membership function (b) Center sensor membership function
motors on the robot body’s right and left. The process carried
out in the fuzzy inference system can be seen in Fig. 3.

(c) Right sensor membership function (d) Left speed membership function

Fig. 3. Functional block of fuzzy inference system

Based on Fig. 3, crisp input will be converted to fuzzy form

through fuzzification. In fuzzification, the membership of the
function of the distance variable is arranged based on three
parts: near, medium, and far. The distance variable is obtained
from ultrasonic sensor readings on the right, middle, and left. (e) Right speed membership function
The membership function equation for each sensor can be Fig. 4. Membership function of each variable in (1)-(5)
seen in Equations (1)-(3). Fig. 4 (a)-(e) shows the membership
function of the distance and speed variable. Each input consists DC motor speed control is carried out to avoid obstacles in
of three set functions: near, medium, and far. The membership front of the DDMR. The membership function of the speed
functions for the left, center and right sensors are the same. variable in each DC motor can be seen in Equations (4)-(5).
Fig. 4(b) shows the membership function of the speed variable.

 
1
 if xi < 5 1
 if zj < −50
12−xi zj
µnear (xi ) = if 5 ≤ xi ≤ 12 (1) µnegative (zj ) = − 50 if − 50 ≤ zj ≤ 0 (4)
 7 
0 if x > 12 0 if zj > 0
 

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 137

IV. R ESULT AND D ISCUSSION


0
 if zj < 0 The initial test results are mathematical calculations to obtain
zj
µpositive (zj ) = if 0 ≤ zj ≤ 50 (5) each rule’s left and right motor speed values. The left, center,
 50
0 if zj > 50 and right sensor values represent each condition in the rule.

Table II shows the speed values for the left and right motor
where j is left and right motor. speeds under various conditions. When the left, middle, and
Based on Fig. 4, the fuzzy rules that will be used for the right sensor conditions are 6 cm, 6 cm, and 7 cm, respectively,
robot’s obstacle avoidance system are listed in Table I. Rules calculating the left and right motor speed values is carried out
are the link between input and output variables. All rules 1 to in several stages, according to Fig. 3. The steps involved in
27 in the table are obtained heuristically, where these rules are getting the right and left motor speed are explained below.
used as a reference for the robot’s output movement on the DC TABLE II. S UMMARY OF MATHEMATICAL CALCULATIONS
motors.
Rule Ultrasonic Sensor DC Motor
TABLE I. RULES OF F UZZY I NFENECE S YSTEM
Left Center Right Left Right
1 6 6 9 -16 -50
Rule Ultrasonic Sensor DC Motor 2 5 6 20 36 -36
Left Center Right Left Right 3 6 15 7 35 35
1 near near near negative negative 4 6 16 20 36 -36
2 near near medium positive negative 5 17 7 6 -35 35
3 near medium near positive positive 6 20 20 6 -36 36
4 near medium medium positive negative 7 14 5 15 36 -36
5 medium near near negative positive 8 15 20 25 37 37
6 medium medium near negative positive 9 6 5 40 36 -36
7 medium near medium positive negative 10 5 40 5 37 37
8 medium medium medium positive positive 11 7 39 41 35 -35
9 near near far positive negative 12 38 5 6 -35 35
10 near far near positive positive 13 39 40 6 -36 36
11 near far far positive negative 14 40 5 40 37 -37
12 far near near negative positive 15 38 39 40 35 35
13 far far near negative positive 16 20 39 45 36 36
14 far near far positive negative 17 17 39 36 50 50
15 far far far positive positive 18 15 15 39 36 36
16 medium far far positive positive 19 39 36 15 50 50
17 medium far medium positive positive 20 40 15 40 37 37
18 medium medium far positive positive 21 39 20 15 36 36
19 far far medium positive positive 22 7 41 17 35 -35
20 far medium far positive positive 23 9 15 39 50 -16
21 far medium medium positive positive 24 39 7 20 -35 35
22 near far medium positive negative 25 40 15 6 -36 36
23 near medium far positive negative 26 15 7 36 50 -50
24 far near medium negative positive 27 15 40 5 -37 37
25 far medium near negative positive
26 medium near far positive negative
27 medium far near negative positive
1) Fuzzification
In this case, the value of the fuzzy set is
The next step is aggregation, combining the IF-THEN rule
output into a single fuzzy set. In this research, it was determined • Left sensor : µnear (6) = 0.86
using the MIN function to produce a single fuzzy set. The re-
sults of the aggregation process are still fuzzy information. For • Center sensor : µnear (6) = 0.86
this reason, it is necessary to carry out calculations that produce
a single number as the controller output value (defuzzification). • Right sensor : µnear (9) = 0.43, µmedium (9) = 0.14
The defuzzification process is carried out to get crisp values
from fuzzy values. The defuzzification process in the research 2) Rule of fuzzy inference system (MINIMUM)
uses the Center of Gravity (COG) method, with appropriate For left motor speed:
calculations in (6).

R • IF Left = near AND Center = near AND Right =

µspeed (zj )zj dz near THEN Left Speed = negative (0.43)
zj∗ = R (6)
µspeed (zj )dz

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 138

• IF Left = near AND Center = near AND Right = Table II, the left and right motor speed values show compliance
medium THEN Left Speed = positive (0.14) with the rules created in Table I.
For right motor speed: These results are then implemented in DDMR to avoid
obstacles in real-time. Implementing fuzzy logic in obstacle
• IF Left = near AND Center = near AND Right = avoidance of DDMR is carried out in the test environment
near THEN Left Speed = negative (0.43) according to Fig. 6. DDMR moves from the start position to
the goal position autonomously.
IF Left = near AND Center = near AND Right =
• The three environments used in testing represent obstacles in
medium THEN Left Speed = negative (0.14) the form of passageways lined with walls. Tests were carried
3) Aggregation using the MAXIMUM rule out to see DDMR’s ability to avoid obstacles, especially in
For left motor speed: producing turning left and right maneuvers.


 0.43 if − 55 ≤ zl ≤ −21.5
− 0.43zl

if − 21.5 < zl ≤ 0
zl = 0.14z21.5 (7)


 7
l
if 0 < zl ≤ 7
0.14 if 7 < zl ≤ 55


For right motor speed:

(
0.43 if − 55 ≤ zl ≤ −21.5
zr = (8)
− 0.43z
21.5
l
if − 21.5 < zl ≤ 0 (a) Environment 1 (b) Environment 2

Representation of aggregation according to (7) and (8)

can be seen in Fig. 5.

(c) Environment 3

Fig. 6. Environment of the test

The results of DDMR testing in three environments can be

(a) Left speed (b) Right speed seen in Fig. 7. Based on Fig. 7, DDMR can move towards
Fig. 5. Membership function of each variable in (1)-(5) the goal position without hitting surrounding obstacles. In
environment 1, DDMR can reach the goal position in 101 s.
4) Defuzzification The distance DDMR travels to get the goal position from
−550.99 + 66.26 + 2.29 + 208.32 the start position is 4.7 m. In environments 2 and 3, DDMR
∗
zlef t = = −16rpm reached the target without hitting surrounding obstacles. The
14.41 + (−4.62) + 0.49 + 6.72
DDMR travel distance in environments 2 and 3 is 5.0 m and
∗ −550.99 + 66.26 5.5 m, respectively. The time required for DDMR to reach the
zright = = −50rpm
14.41 + (−4.62) goal position in environments 2 and 3 is 107 s and 102 s,
respectively. The test summary results can be seen in Table III.
The left and right motor speed values can be seen from
The average speed produced by DDMR in reaching the target
the defuzzification process. The speed values of the left motor
∗ ∗ is 4.91 cm/s.
(zlef t ) and right motor (zright ) are -16 rpm and -50 rpm, re-
spectively. This shows that when the sensor detects an obstacle TABLE III. R ESULT OF THE TEST
close to the DDMR, the speed of the right and left wheels
will be negative. Therefore, in that condition, DDMR moves Environment Distance Travelled Travelling Time
(cm) (s)
backwards. 1 470 101
Mathematical analysis is carried out on all rules using case 2 500 107
3 550 102
examples. The results of mathematical calculations can be seen
in Table II. Based on the results of mathematical calculations in

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 139

[6] M. Cardona, F. Cortez, A. Palacios, and K. Cerros, “Mobile robots

Application Against Covid-19 Pandemic,” in 2020 IEEE ANDESCON,
pp. 1–5, 2020 doi: 10.1109/ANDESCON50619.2020.9272072.
[7] B. K. Patle, G. Babu L, A. Pandey, D. R. K. Parhi, and A. Ja-
gadeesh, “A review: On path planning strategies for navigation of mobile
robot,” Defence Technology, vol. 15, no. 4, pp. 582–606, 2019, doi:
10.1016/j.dt.2019.04.011.
[8] T. Kot and P. Novák, “Application of virtual reality in teleoperation of
the military mobile robotic system TAROS,” International journal of
advanced robotic systems, vol. 15, no. 1, p. 1729881417751545, 2018,
(a) Result 1 (b) Result 2 doi: 10.1177/1729881417751545.
[9] Z. Luo, J. Shang, G. Wei, and L. Ren, “A reconfigurable hy-
brid wheel-track mobile robot based on Watt II six-bar linkage,”
Mechanism and Machine Theory, vol. 128, pp. 16–32, 2018, doi:
10.1016/j.mechmachtheory.2018.04.020.
[10] F. Arvin, J. Espinosa, B. Bird, A. West, S. Watson, and B. Lennox, “Mona:
an Affordable Open-Source Mobile robot for Education and Research,”
Journal of Intelligent & Robotic Systems, vol. 94, no. 3–4, pp. 761–775,
2019, doi: 10.1007/s10846-018-0866-9.
[11] M. Ben-Ari and F. Mondada, “Robots and Their Applications,” in
Elements of Robotics, Cham: Springer International Publishing, pp. 1–20,
(c) Result 3 2018 doi: 10.1007/978-3-319-62533-1 1.
[12] J. M. Cañas, E. Perdices, L. Garcı́a-Pérez, and J. Fernández-Conde,
Fig. 7. DDMR trajectory in each environment “A ROS-Based Open Tool for Intelligent Robotics Education,” Applied
Sciences, vol. 10, no. 21, p. 7419, 2020, doi: 10.3390/app10217419.
[13] X. Wu et al., “The autonomous navigation and obstacle avoidance
V. C ONCLUSION for USVs with ANOA deep reinforcement learning method,”
Knowledge-Based Systems, vol. 196, p. 105201, 2020, doi:
10.1016/j.knosys.2019.105201.
This research implements the Mamdani fuzzy inference sys- [14] F. H. Ajeil, I. K. Ibraheem, A. T. Azar, and A. J. Humaidi, “Autonomous
tem on a low-cost Differential Drive Mobile Robot (DDMR). navigation and obstacle avoidance of an omnidirectional mobile robot
using swarm optimization and sensors deployment,” nternational Journal
Three ultrasonic sensors arranged at an angle of 450 can identify of Advanced Robotic Systems, vol. 17, no. 3, p. 1729881420929498, 2020,
obstacles in front of the DDMR. The variable distance from doi: 10.1177/1729881420929498.
the three sensors was successfully used to regulate the speed of [15] M. Rubagotti, T. Taunyazov, B. Omarali, and A. Shintemirov, “Semi-
the left and right DDMR motors. Based on testing, DDMR can Autonomous Robot Teleoperation With Obstacle avoidance via Model
Predictive Control,” IEEE Robotics and Automation Letters, vol. 4, no. 3,
avoid obstacles and move towards the target. DDMR kinematics pp. 2746–2753, 2019, doi: 10.1109/LRA.2019.2917707.
can be added for further research to produce more natural [16] N. H. Singh and K. Thongam, “Mobile robot Navigation Using Fuzzy
movements. logic in Static Environments,” Procedia Computer Science, vol. 125,
pp. 11–17, 2018, doi: 10.1016/j.procs.2017.12.004.
This research was carried out by forming an angle in the [17] A. Shitsukane, W. Cheruiyot, C. Otieno, and M. Mvurya, “Fuzzy logic
testing environment with a considerable angle value. In the next Sensor Fusion for Obstacle avoidance Mobile robot,” in 2018 IST-Africa
Week Conference, pp. 1-8, 2018.
test, testing in an environment with a narrow corner trap is
[18] A. Nasrinahar and J. H. Chuah, “Intelligent motion planning of a mobile
necessary. A controller is also needed to regulate the robot’s robot with dynamic obstacle avoidance,” Journal on Vehicle Routing
speed to reach the desired speed. Algorithms, vol. 1, no. 2–4, pp. 89–104, 2018, doi: 10.1007/s41604-018-
0007-4.
[19] M. Nadour, M. Boumehraz, L. Cherroun, and V. Puig, “Hybrid Type-2
R EFERENCES Fuzzy logic Obstacle avoidance System based on Horn-Schunck Method,”
Electrotehnica, Electronica, Automatica, vol. 67, no. 3, pp. 45-51, 2019.
[1] M. Javaid, A. Haleem, R. P. Singh, and R. Suman, “Substantial capa- [20] J. Zheng, B. Liu, Z. Meng, and Y. Zhou, “Integrated real time obstacle
bilities of robotics in enhancing industry 4.0 implementation,” Cognitive avoidance algorithm based on fuzzy logic and L1 control algorithm for
Robotics, vol. 1, pp. 58–75, 2021, doi: 10.1016/j.cogr.2021.06.001. unmanned helicopter,” in 2018 Chinese Control And Decision Conference,
[2] G. Fragapane, D. Ivanov, M. Peron, F. Sgarbossa, and J. O. Strandhagen, pp. 1865–1870, 2018 doi: 10.1109/CCDC.2018.8407430.
“Increasing flexibility and productivity in Industry 4.0 production net- [21] E. Pourjavad and R. V. Mayorga, “A comparative study and measur-
works with autonomous mobile robots and smart intralogistics,” Annals ing performance of manufacturing systems with Mamdani fuzzy infer-
of operations research, vol. 308, no. 1–2, pp. 125–143, Jan. 2022, doi: ence system,” Journal of Intelligent Manufacturing, vol. 30, no. 3,
10.1007/s10479-020-03526-7. pp. 1085–1097, 2019, doi: 10.1007/s10845-017-1307-5.
[3] F. D’Souza, J. Costa, and J. N. Pires, “Development of a solution for [22] H. Rawat, D. R. Parhi, P. B. Kumar, K. K. Pandey, and A. K. Behera,
adding a collaborative robot to an industrial AGV,” Industrial Robot: “Analysis and Investigation of Mamdani Fuzzy for Control and Navigation
The International Journal of Robotics Research and Applicationl, vol. 47, of Mobile robot and Exploration of different AI techniques pertaining
no. 5, pp. 723–735, 2020, doi: 10.1108/IR-01-2020-0004. to Robot Navigation,” Emerging trends in Engineering, Science and
[4] J. Holland et al., “Service Robots in the Healthcare Sector,” Robotics, Manufacturing, 2018.
vol. 10, no. 1, p. 47, 2021, doi: 10.3390/robotics10010047. [23] A. Najmurrokhman, U. Komarudin, E. C. Djamal, and F. Taufik, “Speed
[5] F. Zhou, X. Wang, and M. Goh, “Fuzzy extended VIKOR-based mobile control and obstacle avoidance of a hexapod mobile robot using mam-
robot selection model for hospital pharmacy,” International Journal of dani type fuzzy logic controller,” in 2019 6th International Conference
Advanced Robotic Systems, vol. 15, no. 4, p. 1729881418787315, 2018, on Instrumentation, Control, and Automation, pp. 199–202, 2019, doi:
doi: 10.1177/1729881418787315. 10.1109/ICA.2019.8916714.

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 140

[24] M. Shen, Y. Wang, Y. Jiang, H. Ji, B. Wang, and Z. Huang, “A Path Planning Algorithms for Mobile Robots,” Vehicles, vol. 3, no. 3,
New Positioning Method Based on Multiple Ultrasonic Sensors for pp. 448–468, 2021, doi: 10.3390/vehicles3030027.
Autonomous Mobile robot,” Sensors, vol. 20, no. 1, p. 17, 2019, doi: [43] Y. R. Petillot, G. Antonelli, G. Casalino, and F. Ferreira, “Underwater
10.3390/s20010017. Robots: From Remotely Operated Vehicles to Intervention-Autonomous
[25] X. Chen, S. Wang, B. Zhang, and L. Luo, “Multi-feature fusion tree trunk Underwater Vehicles,” in IEEE Robotics & Automation Magazine, vol. 26,
detection and orchard mobile robot localization using camera/ultrasonic no. 2, pp. 94–101, 2019, doi: 10.1109/MRA.2019.2908063.
sensors,” Computers and Electronics in Agriculture, vol. 147, pp. 91–108, [44] C. J. Munoz Martinez, R. Castro Salguero, R. Palomares, and J. Cornejo,
2018, doi: 10.1016/j.compag.2018.02.009. “Mechatronics Development of Terrestrial Mobile Robot for Exploring
[26] M. Derkach, D. Matiuk, and I. Skarga-Bandurova, “Obstacle avoid- and Monitoring Environmental Parameters at Mine Analogue Sites using
ance Algorithm for Small Autonomous Mobile robot Equipped with IoT Platform,” in 2020 IEEE XXVII International Conference on Elec-
Ultrasonic Sensors,” in 2020 IEEE 11th International Conference on tronics, Electrical Engineering and Computing (INTERCON), pp. 1–4,
Dependable Systems, Services and Technologies, pp. 236–241, 2020, doi: 2020, doi: 10.1109/INTERCON50315.2020.9220227.
10.1109/DESSERT50317.2020.9125019. [45] C. Silva, Á. De Oliveira, and M. Fernandes, “Validation of a Dynamic
[27] Y. Cheng and G. Y. Wang, “Mobile robot navigation based on lidar,” in Planning Navigation Strategy Applied to Mobile Terrestrial Robots,”
2018 Chinese Control And Decision Conference, pp. 1243–1246, 2018, Sensors, vol. 18, no. 12, p. 4322, 2018, doi: 10.3390/s18124322.
doi: 10.1109/CCDC.2018.8407319. [46] X. Zhang et al., “Large-scale 3D printing by a team of mobile
[28] D. Hutabarat, M. Rivai, D. Purwanto, and H. Hutomo, “Lidar-based robots,” Automation in Construction, vol. 95, pp. 98–106, 2018, doi:
Obstacle avoidance for the Autonomous Mobile robot,” in 2019 12th 10.1016/j.autcon.2018.08.004.
International Conference on Information & Communication Technology [47] H. Huang, A. V. Savkin, M. Ding, and C. Huang, “Mobile robots
and System, pp. 197–202, 2019, doi: 10.1109/ICTS.2019.8850952. in wireless sensor networks: A survey on tasks,” Computer Networks,
[29] F. Rubio, F. Valero, and C. Llopis-Albert, “A review of mobile vol. 148, pp. 1–19, 2019, doi: 10.1016/j.comnet.2018.10.018.
robots: Concepts, methods, theoretical framework, and applications,” [48] M. B. Alatise and G. P. Hancke, “A Review on Challenges of Autonomous
International Journal of Advanced Robotic Systems, vol. 16, no. 2, Mobile Robot and Sensor Fusion Methods,” in IEEE Access, vol. 8,
p. 1729881419839596, 2019, doi: 10.1177/1729881419839596. pp. 39830–39846, 2020, doi: 10.1109/ACCESS.2020.2975643.
[30] X. Liang, H. Wang, Y.-H. Liu, W. Chen, and Z. Jing, “Image-Based [49] X. Gao et al., “Review of Wheeled Mobile Robots’ Navigation Problems
Position Control of Mobile Robots With a Completely Unknown Fixed and Application Prospects in Agriculture,” in IEEE Access, vol. 6,
Camera,” in IEEE Transactions on Automatic Control, vol. 63, no. 9, pp. 49248–49268, 2018, doi: 10.1109/ACCESS.2018.2868848.
pp. 3016–3023, 2018, doi: 10.1109/TAC.2018.2793458. [50] B. Song, Z. Wang, and L. Zou, “An improved PSO algorithm for
[31] Z. Feng, G. Hu, Y. Sun, and J. Soon, “An overview of collaborative robotic smooth path planning of mobile robots using continuous high-degree
manipulation in multi-robot systems,” Annual Reviews in Control, vol. 49, Bezier curve,” Applied Soft Computing, vol. 100, p. 106960, 2021, doi:
pp. 113–127, 2020, doi: 10.1016/j.arcontrol.2020.02.002. 10.1016/j.asoc.2020.106960.
[32] S. Chang, Y. Wang, and Z. Zuo, “Fixed-Time Active Disturbance Re- [51] A. Stefek, T. V. Pham, V. Krivanek, and K. L. Pham, “Energy Comparison
jection Control and Its Application to Wheeled Mobile Robots,” in IEEE of Controllers Used for a Differential Drive Wheeled Mobile Robot,”
Transactions on Systems, Man, and Cybernetics: Systems, vol. 51, no. 11, in IEEE Access, vol. 8, pp. 170915-170927, 2020, doi: 10.1109/AC-
pp. 7120–7130, 2021, doi: 10.1109/TSMC.2020.2966077. CESS.2020.3023345.
[52] X. Wu, P. Jin, T. Zou, Z. Qi, H. Xiao, and P. Lou, “Backstepping
[33] N. Kousi, C. Gkournelos, S. Aivaliotis, C. Giannoulis, G. Michalos, and
Trajectory Tracking Based on Fuzzy Sliding Mode Control for Differential
S. Makris, “Digital twin for adaptation of robots’ behavior in flexible
Mobile Robots,” Journal of Intelligent & Robotic Systems, vol. 96, no. 1,
robotic assembly lines,” Procedia Manufacturing, vol. 28, pp. 121–126,
pp. 109–121, 2019, doi: 10.1007/s10846-019-00980-9.
2019, doi: 10.1016/j.promfg.2018.12.020.
[53] T. Mylvaganam and M. Sassano, “Autonomous collision avoidance
[34] S. M. H. Rostami, A. K. Sangaiah, J. Wang, and X. Liu, “Obstacle avoid-
for wheeled mobile robots using a differential game approach,”
ance of mobile robots using modified artificial potential field algorithm,”
European Journal of Control, vol. 40, pp. 53–61, 2018, doi:
J Wireless Com Network, vol. 70, 2019, doi: 10.1186/s13638-019-1396-2.
10.1016/j.ejcon.2017.11.005.
[35] G. Hoffman, “Evaluating Fluency in Human–Robot Collaboration,” [54] A. Štefek, V. T. Pham, V. Krivanek, and K. L. Pham, “Optimization
in IEEE Transactions on Human-Machine Systems, vol. 49, no. 3, of Fuzzy Logic Controller Used for a Differential Drive Wheeled Mo-
pp. 209–218, 2019, doi: 10.1109/THMS.2019.2904558. bile Robot,” Applied Sciences, vol. 11, no. 13, p. 6023, 2021, doi:
[36] N. Boysen, S. Schwerdfeger, and F. Weidinger, “Scheduling last-mile 10.3390/app11136023.
deliveries with truck-based autonomous robots,” European Journal of [55] D. An, S. Ji, and I. U. Jan, “Investigating the Determinants and Barriers of
Operational Research, vol. 271, no. 3, pp. 1085–1099, 2018, doi: Purchase Intention of Innovative New Products,” Sustainability, vol. 13,
10.1016/j.ejor.2018.05.058. no. 2, p. 740, 2021, doi: 10.3390/su13020740.
[37] M. Gadaleta, M. Pellicciari, and G. Berselli, “Optimization of the energy [56] M. Bikova, V. O. Latkoska, B. Hristov, and D. Stavrov, “Path Plan-
consumption of industrial robots for automatic code generation,” Robotics ning Using Fuzzy Logic Control of a 2-DOF Robotic Arm,” in 2022
and Computer-Integrated Manufacturing, vol. 57, pp. 452–464, 2019, doi: IEEE 17th International Conference on Control & Automation (ICCA),
10.1016/j.rcim.2018.12.020. pp. 998–1003, 2022 doi: 10.1109/ICCA54724.2022.9831903.
[38] A. Le, V. Prabakaran, V. Sivanantham, and R. Mohan, “Modified A-Star [57] C.-C. Chang, C. Liang, and Y.-C. Chiu, “Direct or indirect effects from
Algorithm for Efficient Coverage Path Planning in Tetris Inspired Self- ‘perceived characteristic of innovation’ to ‘intention to pay’: mediation
Reconfigurable Robot with Integrated Laser Sensor,” Sensors, vol. 18, of continuance intention to use e-learning,” Journal of Computers in
no. 8, p. 2585, 2018, doi: 10.3390/s18082585. Education, vol. 7, no. 4, pp. 511–530, 2020, doi: 10.1007/s40692-020-
[39] A. Krishna Lakshmanan et al., “Complete coverage path planning using 00165-6.
reinforcement learning for Tetromino based cleaning and maintenance [58] P. Duraisamy, M. N. Santhanakrishnan, and R. Amirtharajan, “Genetic
robot,” Automation in Construction, vol. 112, p. 103078, 2020, doi: Algorithm Optimized Grey-Box Modelling and Fuzzy Logic Controller
10.1016/j.autcon.2020.103078. for Tail-Actuated Robotic Fish,” Neural Processing Letters, vol. 55,
[40] Z. Wang, H. Li, and X. Zhang, “Construction waste recycling robot pp. 11577-11594, 2023, doi: 10.1007/s11063-023-11391-1.
for nails and screws: Computer vision technology and neural network [59] C. Dong, Z. Yu, X. Chen, H. Chen, Y. Huang, and Q. Huang, “Adapt-
approach,” Automation in Construction, vol. 97, pp. 220–228, 2019, doi: ability Control Towards Complex Ground Based on Fuzzy Logic for
10.1016/j.autcon.2018.11.009. Humanoid Robots,” in IEEE Transactions on Fuzzy Systems, vol. 30,
[41] Mohd. N. Zafar and J. C. Mohanta, “Methodology for Path Planning and no. 6, pp. 1574–1584, 2022, doi: 10.1109/TFUZZ.2022.3167458.
Optimization of Mobile Robots: A Review,” Procedia Computer Science, [60] A. D. Hayat, A. Hassan, M. Hajar, and E. A. Ali, “Fuzzy Logic
vol. 133, pp. 141–152, 2018, doi: 10.1016/j.procs.2018.07.018. Controller for 4WD/4WS Autonomous Agricultural Robotic,” in 2022
[42] K. Karur, N. Sharma, C. Dharmatti, and J. E. Siegel, “A Survey of IEEE 3rd International Conference on Electronics, Control, Optimization

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot
Journal of Robotics and Control (JRC) ISSN: 2715-5072 141

and Computer Science (ICECOCS), pp. 1–6, 2022 doi: 10.1109/ICE- [76] M. R. A. Refaai, “An Improved Inverse Kinematics Solution for a
COCS55148.2022.9983035. Robot Arm Trajectory Using Multiple Adaptive Neuro-Fuzzy Inference
[61] Z. Hou, Z. Li, C. Hsu, K. Zhang, and J. Xu, “Fuzzy Logic-Driven Vari- Systems,” Advances in Materials Science and Engineering, vol. 2022,
able Time-Scale Prediction-Based Reinforcement Learning for Robotic pp. 1–12, 2022, doi: 10.1155/2022/1413952.
Multiple Peg-in-Hole Assembly,” in IEEE Transactions on Automation [77] Md. R. Sarkar, Mst. J. Nahar, A. Nadia, Md. A. Halim, S. M. S.
Science and Engineering, vol. 19, no. 1, pp. 218–229, Jan. 2022, doi: Hossain Rafin, and Md. M. Rahman, “Proficiency Assessment of Adaptive
10.1109/TASE.2020.3024725. Neuro-Fuzzy Inference System to Predict Wind Power: A Case Study of
[62] M. R. Mohd Romlay, A. Mohd Ibrahim, S. F. Toha, P. De Wilde, I. Malaysia,” in 2019 1st International Conference on Advances in Science,
Venkat, and M. S. Ahmad, “Obstacle avoidance for a robotic navigation Engineering and Robotics Technology (ICASERT), pp. 1–5, 2019, doi:
aid using Fuzzy Logic Controller-Optimal Reciprocal Collision Avoidance 10.1109/ICASERT.2019.8934557.
(FLC-ORCA),”Neural Computing and Applications, vol. 35, no. 30, [78] P. Subbash and K. T. Chong, “Adaptive network fuzzy inference system
pp. 22405–22429, 2023, doi: 10.1007/s00521-023-08856-8. based navigation controller for mobile robot,” Frontiers of Information
[63] G. S. Maraslidis, T. L. Kottas, M. G. Tsipouras, and G. F. Frag- Technology & Electronic Engineering, vol. 20, no. 2, pp. 141–151, 2019,
ulis, “Design of a Fuzzy Logic Controller for the Double Pendulum doi: 10.1631/FITEE.1700206.
Inverted on a Cart,” Information, vol. 13, no. 8, p. 379, 2022, doi: [79] R. Suppiah, N. Kim, A. Sharma, and K. Abidi, “Fuzzy inference system
10.3390/info13080379. (FIS) - long short-term memory (LSTM) network for electromyography
[64] C. Ntakolia, K. S. Platanitis, G. P. Kladis, C. Skliros, and A. D. (EMG) signal analysis,” Biomedical Physics & Engineering Express,
Zagorianos, “A Genetic Algorithm enhanced with Fuzzy-Logic for vol. 8, no. 6, p. 065032, 2022, doi: 10.1088/2057-1976/ac9e04.
multi-objective Unmanned Aircraft Vehicle path planning missions,” in [80] K. Takacs and T. Haidegger, “Adaptive Neuro-fuzzy Inference System
2022 International Conference on Unmanned Aircraft Systems (ICUAS), for Automated Skill Assessment in Robot-Assisted Minimally Inva-
pp. 114–123. 2022, doi: 10.1109/ICUAS54217.2022.9836068. sive Surgery,” in 2021 IEEE 25th International Conference on In-
telligent Engineering Systems (INES), pp. 000215–00022, 2021, doi:
[65] A. M. Serifoglu and C. Kasnakoglu, “Fuzzy Logic Controlled Active
10.1109/INES52918.2021.9512924.
Hydro-Pneumatic Suspension Design Simulation and Comparison for
Performance Analysis,” in 2021 3rd International Congress on Human-
Computer Interaction, Optimization and Robotic Applications (HORA),
pp. 1–5, 2021, doi: 10.1109/HORA52670.2021.9461373.
[66] C. F. Riman and P. E. Abi-Char, “Fuzzy Logic Control for Mobile Robot
Navigation in Automated Storage,” nternational Journal of Mechanical
Engineering and Robotics Research, vol. 12, no. 5, pp. 313–323, 2023,
doi: 10.18178/ijmerr.12.5.313-323.
[67] I. A. B K, A. Prakash, and A. Prakash, “A Robo-Fuzzy Inference System
for Iron based Sprouts Fortification,” in 2021 Second International Con-
ference on Electronics and Sustainable Communication Systems (ICESC),
pp. 463–468, 2021, doi: 10.1109/ICESC51422.2021.9532621.
[68] E. Brumancia, S. Justin Samuel, L. M. Gladence, and K. Rathan, “Hybrid
data fusion model for restricted information using Dempster–Shafer
and adaptive neuro-fuzzy inference (DSANFI) system,” Soft Computing,
vol. 23, no. 8, pp. 2637–2644, 2019, doi: 10.1007/s00500-018-03734-1.
[69] X. Chen, Y. Leng, and C. Fu, “A Supervised-Reinforced Successive
Training Framework for a Fuzzy Inference System and Its Application
in Robotic Odor Source Searching,” Frontiers in Neurorobotics, vol. 16,
p. 914706, 2022, doi: 10.3389/fnbot.2022.914706.
[70] P. Dutta and N. Anjum, “Optimization of Temperature and Relative
Humidity in an Automatic Egg Incubator Using Mamdani Fuzzy Inference
System,” in 2021 2nd International Conference on Robotics, Electrical
and Signal Processing Techniques (ICREST), pp. 12–16, 2021, doi:
10.1109/ICREST51555.2021.9331155.
[71] R. Gimazov and S. Shidlovskiy, “The architecture of adaptive neural
network based on a fuzzy inference system for implementing intelligent
control in photovoltaic systems,” IOP Conference Series: Materials Sci-
ence and Engineering, vol. 363, p. 012016, 2018, doi: 10.1088/1757-
899X/363/1/012016.
[72] M. H. Haider et al., “Autonomous Mobile Robot Navigation using Adap-
tive Neuro Fuzzy Inference System,” in 2022 International Conference on
Innovations and Development of Information Technologies and Robotics
(IDITR), pp. 93–99, 2022 doi: 10.1109/IDITR54676.2022.9796495.
[73] M. Mukhtar, D. Khudher, and T. Kalganova, “A control structure for
ambidextrous robot arm based on Multiple Adaptive Neuro-Fuzzy In-
ference System,” IET Control Theory & Applications, vol. 15, no. 11,
pp. 1518–1532, 2021, doi: 10.1049/cth2.12140.
[74] G. Özden, M. Ö. Öteyaka, and F. M. Cabrera, “Modeling of cutting
parameters in turning of PEEK composite using artificial neural networks
and adaptive-neural fuzzy inference systems,” Journal of Thermoplastic
Composite Materials, vol. 36, no. 2, pp. 493–509, Feb. 2023, doi:
10.1177/08927057211013070.
[75] M. R. A. Refaai, “Using Multiple Adaptive Neuro-Fuzzy Inference
System to Solve Inverse Kinematics of SCARA Robot,” in 2021 18th
International Multi-Conference on Systems, Signals & Devices (SSD), pp.
154–159, 2021, doi: 10.1109/SSD52085.2021.9429498.

Riky Dwi Puriyanto, Design and Implementation of Fuzzy Logic for Obstacle Avoidance in Differential Drive Mobile Robot