Design of Ground Vehicle System Semi-autonomous Preceder Type

for Straight Path and Circular Path using Fuzzy Logic Method

Illa Rizianiza, Adhe Yusphie Panca Tentra Sandika

Institut Teknologi Kalimantan

Keywords: Fuzzy Logic, Ground Vehicle, Semi-Autonomous.

Abstract: The development of the robotics system increase significantly. Recently, robotics become an opportunity for

people to use those in many activities, for instance, automatic control systems in the land, which can be

referred to as Unmanned Ground Vehicle (UGV). Unmanned Ground Vehicle (UGV) has several categories

based on autonomous systems provided, namely teleoperated, semi-autonomous, platform-centric, and

network-centric. In this study, the prototype of the UGV system is a semi-autonomous type using the fuzzy

logic method as a control system. The input of fuzzy logic is the distance between the prototype and

obstacle. However, the output variable is the action of UGV to control the safe distance from the obstacle.

The design fuzzy logic system has been integrated with the robotdyn UNO microcontroller device as a

controller of UGV. The test results showed that the prototype could keep tracking to control the distance

according to rule base fuzzy logic. The setpoint has been set up at 10 cm - 15 cm from the obstacle and the

results of the semi-autonomous system movement that has been applied in-ground vehicle prototype. The

average error is 9 cm. This number still intolerance in tracking control.

1 INTRODUCTION

The autonomous ground vehicle has a wide scope of

research consisting of the short, medium, and long-

range robots that have small and large sizes with the

aim of supervision in small sectors, surveys, and

others. In vehicles that have a large size, it is usually

controlled by a combination of artificial intelligence

and human operators at a considerable distance. So

far in Indonesia, the land vehicle movement system

is still fully controlled by humans with the value of

Human-Robot Interaction (HRI) of 100% or level 0,

especially in fields where the land has bad

conditions such as bumpy, uneven, there are many

obstacles and others. The autonomous ground

vehicle has profound theoretical-practical significant

in an intelligent transport system (Kiencke, 2001). In

autonomous ground vehicle technologies,

longitudinal control keeps it moving at the desired

speed by controlling throttle and brake coordinately.

However, the non-linearity and uncertainty in the

dynamic model of the vehicle introduce difficulties

in the design of a longitudinal controller with high

precision (Xiaolon, 2011).

In this paper, as an extension of an autonomous

ground vehicle designed and developed for the user,

the control system by fuzzy logic presented. This

vehicle used a fuzzy control system to the desired

track. Fuzzy logic has the ability to transform human

sematic processes into numerical machine processes.

Fuzzy logic provides a link between human intuition

and machine expression. Fuzzy control methods are

known as powerful, intelligent tools that can be used

for controlling complex nonlinear systems.

Performances of fuzzy controllers have been

presented in commercial products and industrial

control applications (Kim, 2013). In this paper, a

fuzzy control system has simulated to the straight

and circular path.

2 FUZZY LOGIC CONTROL

2.1 Fuzzy Logic

Fuzzy logic was first discovered by Prof. Lotfi

Asker Zadeh at a seminar at the University of

California, Berkeley, United States. Fuzzy logic is

made on the grounds that there is uncertainty in

which members cannot only be separated into

Rizianiza, I. and Sandika, A.

Design of Ground Vehicle System Semi-autonomous Preceder Type for Straight Path and Circular Path using Fuzzy Logic Method.

DOI: 10.5220/0009443401770183

In Proceedings of the 1st International Conference on Industrial Technology (ICONIT 2019), pages 177-183

ISBN: 978-989-758-434-3

177

members (1) or not members (0). There are several

definitions of fuzzy logic, including:

1. Fuzzy logic is a logic used to explain

ambiguity and explain the logic of a set that

resolves the ambiguity.

2. Fuzzy logic provides a way to convert

linguistic statements into a numeric (Hirulkar,

2014).

Figure 1: Fuzzy Controller

Fuzzy logic systems, which can be referred to as

Fuzzy Logic Control, have a composition of the

main setting components consisting of fuzzification,

rule base, mechanism of inference, and

defuzzification with each Fuzzy Logic Control

having input and output (Maghfiroh, 2016).

2.2 Membership Function Fuzzy Logic

Controller

Fuzzy logic also has a membership function, namely

a curve that shows the mapping of input data points

into the value of its membership. There are several

types of curves used to define membership

functions, namely Triangle Membership Function

Curve The function of triangle membership is

determined by three parameters, namely {a, b, c} by

following the rules as follows:

(1)

Figure 2: Triangle Membership Functions

Parameters {a, b, c} with a < b < c determine the

x coordinates of the three angles of the triangle

membership function (Franck, 2013). The Mamdani

method is often also known as the Max-Min

Method. To get the output, four steps are needed:

1. Fuzzy Set Formation

2. Application Function Implications

3. Component Rule

4. Affirmation or Defuzzy

The input from the defuzzification process is a

fuzzy set that is obtained from the composition of

fuzzy rules, while the resulting output is a number in

the fuzzy set domain, so if given a fuzzy set in a

certain range, a certain crisp value must have been

taken as the output. There are several defuzzification

methods in the composition of Mamdani rules,

including the COA method, the bisector, MOM,

LOM, and SOM.

ICONIT 2019 - International Conference on Industrial Technology

178

1. COA Method; In this method, the crisp

solution is obtained by taking the center of the

fuzzy region.

2. Bisector Method; In this method, the crisp

solution is obtained by taking values in a

fuzzy domain, which has a half membership

value from the total membership value in the

fuzzy area.

3. MOM; In this method, the crisp solution is

obtained by taking an average value of the

domain that has the maximum membership

value.

4. LOM; In this method, the crisp solution is

obtained by taking the largest value from the

domain that has the maximum membership

value.

5. SOM; In this method, the crisp solution is

obtained by taking the smallest value from the

domain that has the maximum membership

value.

Fuzzification is the first phase of fuzzy

calculation, which changes input whose definite

truth value is in the form of fuzzy input in the form

of membership level/level of truth. Thus, this stage

takes crisp values and determine the degree to which

they become a member of each corresponding fuzzy

set (Pramudijanto, 2018). The inference is reasoning

using fuzzy input and fuzzy rules that have been

determined to produce fuzzy output.

A variable is a symbol or word that refers to

something that is not certain in the universe of the

discourse. If it's the discourse universe is a set of

numbers, then the variable is called a numerical

variable, whereas if the discourse universe is a set of

words or terms from everyday language (for

example high, fast, young, etc.), then the variable is

called the linguistic variable. The universe of

discourse or universe of words is the whole value

allowed to operate in a fuzzy variable (

Basjaruddin,

2016)

. The universe words are a set of real numbers

which always increases monotone from left to right

or vice versa. The universe word value can be either

positive or negative numbers.

3 METHODOLOGY

Based on the results of the literature study, system

design has been determined aimed at controlling the

action of Ground Vehicle movements realized in the

form of a simple prototype. The block diagram of

the Ground Vehicle movement control system

design system is shown in figure 3. The design of

the Ground Vehicle movement system consists of 4

actions that will be output when the Ground Vehicle

position with the obstacle is at a safe distance or 10

cm - 15 cm, which stops, goes forward, turns right,

and turns left. The design of the movement system

also consists of three main frameworks, namely,

input (input), process (process), and output (output).

The working method of this system itself is where

the input value given by ultrasonic in the form of

distance value to the obstacle will be forwarded to

the robotdyn UNO microcontroller which has been

integrated with fuzzy logic to be processed by the

applied coding, after that the data from robotdyn

Figure 3: Design System of Controller

UNO has been completed processed will be

forwarded to the output value, which will then give

an order to the motor driver to adjust the DC motor

rotation according to the results of the data that has

been given from the robotdyn UNO. This is different

from the process carried out by GPS where input

originating from the

signal transmission to the satellite will be sent back

to the GPS module which will then be processed by

the robotdyn UNO microcontroller with coding that

has been applied and clearly different from previous

coding (without fuzzy logic), after that the result of

the robotdyn UNO process will be forwarded back

to the GPS module as an output by providing a value

in the form of coordinate position or point of

location of the Ground Vehicle.

This design is the initial design carried out to

find out how the composition of each component

Design of Ground Vehicle System Semi-autonomous Preceder Type for Straight Path and Circular Path using Fuzzy Logic Method

179

that will be used is in accordance with the place or

not.

Ultrasonic Sensors are attached to the robotdyn

UNO with the trigger pin and echo pin configuration

inserted into the analog header pin. The continuous

track wheel used has 2 DC motors as the drive, and

the Motor Driver functions to adjust the speed and

speed of the DC motor. After that, the GPS module

will also be paired according to the analog header

pin specified in the robotdyn UNO separately

supplied by the 11.1 Volt 1000 mAh

LithiumPolymer (LiPo) battery power supply.

The hardware design that has been completed

will be adjusted again with the next coding, which

will be explained in the system design. The

hardware design is also included with the wiring

activity, as shown below

Figure 4: Ground Vehicle

4 RESULTS

The rule base of fuzzy logic is used to process input

values, which afterward will be output values as

commands in the form of decisions. This rule base

also has components in the form of certain

Variables which will then be used in the prototype

system. The variables needed to increase the

accuracy of the programming process are three

inputs named input values of the front direction,

values of the right side direction, and values of the

left side direction.

Table 1 Rule Base Fuzzy Logic Ground Vehicle

ICONIT 2019 - International Conference on Industrial Technology

180

Figure 6: Fuzzy Inference System

The formation of a fuzzy inference system is

based on the input, output, and rule base that has

been determined and compiles it based on the

inference mechanism that has been selected, namely

Mamdani's inference mechanism.

Rule readings on the specified member function, as

in Figure 7, starting from the input, the process with

the inference mechanism to output if it is in

accordance with the design plan, the system can be

applied.

Figure 7: Fuzzy Inference System

The results of the first experimental data have

been reshaped into a graph that provides a clear

description of the results obtained, as shown in

Figure 8.

Figure 8: First Experiment Straight Path Without Obstacle

Data

The graph shown in Figure 8 can explain how

the fuzzy logic system performance is applied

directly based on the data obtained for the first

experiment. The X-axis on the graph shows the

distance between the Ground Vehicle and the

obstacle consisting of 3 sides, namely front, left, and

right, and the Y-axis on the graph showing the

amount of data recorded during the Ground Vehicle

functioning from the initial line to the finish line.

The data lines for the left and right directions do not

have a big difference because the values tend to be

the same, so the lines are also almost parallel, but

this is different from the frontline data, which has a

high initial position in the graph with a large initial

value. The value decreases and is almost parallel

with the other 2 data lines because the Ground

Vehicle position that is getting closer to the

destination point will produce a smaller distance

value with a maximum setpoint value of 10 cm from

the obstacle destination point.

For the second experiment, the graphs formed

based on data can be considered in Figure 9 as

follows:

Figure 9: Second Experiment Straight Path Without

Obstacle Data

In Figure 9 it can also be seen that the flow from

the left, front and right data lines has the same

concept as Figure 8 where the left and right data

lines have values that are not much different and

almost parallel, then for the front line data also has

0.5

1.5

2.5

1357911131517

Distance (m)

Sum of Data

Left

Front

Right

0.5

1.5

2.5

1357911131517

Distance (m)

Sum of Data

Left

Front

Right

Design of Ground Vehicle System Semi-autonomous Preceder Type for Straight Path and Circular Path using Fuzzy Logic Method

181

the large initial value continues to change becomes

smaller due to the reduced distance between the

Ground Vehicle and the obstacle destination point so

that the final value is almost parallel to the other 2

data lines.

For the third experiment in Figure 10, there is a

graph formed based on data which can be reviewed

as follows:

Figure 10: Third Experiment Straight Path Without

Obstacle Data

The graphic concept is shown in Figure 10 also

tends to be the same as the previous two graphs,

namely the data lines in the left and right directions

are almost parallel because the values held are not so

different while the front data line has a large initial

value due to the Ground Vehicle position with

obstacle destination point has a long-distance range

so that the more Ground Vehicle position with the

destination point, then the resulting value is also

getting smaller and smaller until the position of the

data line is almost parallel to the other 2 lines.

The value of the distance generated at the front

for this experiment has a value that tends to be

small, and the variations it has are quite large, with a

range of distances that are also small. The graph

formed can be reviewed in Figure 11 as follows:

Figure 11: First Experiment Circular Path Without

Obstacle Data

The results of the graph on the first trial circular

path without obstruction have conciseness different

from the graph in the previous experiment, which is

located at the position of the left sideline whose

value tends to be consistently large starting from the

initial position to the final position. The values on

the right side and the front side have the same

concept as the previous graph, which is almost

parallel because the values possessed by both have a

difference not so far away.

Then for the second experiment on the circular

path, there is a graph that has been formed, as shown

in Figure 12 as follows:

Figure 12: Second Experiment Circular Path Without

Obstacle Data

The concept in Figure 12 also has a form that

tends to be similar to the previous graph image; that

is, on the left side of the data line, it has a beginning

to end value that tends to be large when compared to

2 data on the right and front sides. The values on the

right and front side also have quite a lot of variations

with the characteristics of the value range that is also

small, but it is noted in the data on the right side that

there is a value below 10 cm which is 9 cm from the

obstacle wall arena. The cause is the same as in the

first trial circular path without obstruction.

Then for the third experiment on the circular

path, there is a graph that has been formed, as shown

in Figure 13 as follows:

Figure 13: Third Experiment Circular Path Without

Obstacle Data

0.5

1.5

2.5

1 3 5 7 9 11 13 15 17

Distance (m)

Sum of Data

Left

Front

Right

0.5

1.5

1357911131517

Distance (m)

Sum of Data

Left

Front

Right

0.5

1.5

1 3 5 7 9 11 13 15 17

Distance (m)

Sum of Data

Left

Front

Right

0.5

1.5

1357911131517

Ddistance (m)

Sum of Data

Left

Front

Right

ICONIT 2019 - International Conference on Industrial Technology

182

The graphical form in Figure 13 has a conceptual

basis, which tends to be the same as the two

previous trial graphs where the left side distance

value has a large value starting from the beginning

of the position to the end of the position. Data values

on the right and front also have values that vary with

a small range but tend to be consistent so that the

data lines shown are also almost parallel starting

from the initial position to the final position.

5 CONCLUSION

The conclusion of research of the fuzzy logic control

implemented in the UGV prototype is the control

system, and the objective was reasonably achieved.

The most important in this research is how to design

the control system using fuzzy logic. There is a lot

of variables that should be involved. However, this

research shows a starting point for more advanced

research on the topic similarly. Throughout this

research, the motion of UGV depends on DC motor

as an actuator, the robotdyn UNO microcontroller

based on the fuzzy logic rule base as the control

system and ultrasonic as a sensor to read the distance

between prototype and obstacle.

REFERENCES

Kiencke, U., 2001. Future perspectives of automotive

control. in Proc. 18th IEEE Instrumentation and

Measurement echnology Conf., Budapest. Hungary,

vol. 3, pp. 1510–1518.

Xiaolon, J., Zhibao, S., Xijun, Z., Gong, J., Yan, J., 2011.

Design of a Fuzzy PID Longitudilan Controller for

Autonomous Ground Vehicle. IEEE. pp.269-273.

Kim, H.W., Jung, S., 2013. Fuzzy Control for Balancing

of a Two-Wheel Transportation Robotic

Vehicle:Experiment Studies.IEEE.

Hirulkar, S., Damle, M., Rathee, V., Hardas, B., 2014.

Design of Aoutomatic Car Breaking System Using

Fuzzy Logic and PID Controller.International

Conference on Electronic Systems, Signal Processing

and Computing Technologies, pp. 413-418.

Franck, D., 2013, Introduction to Fuzzy Logic,

Massachusetts Institute of Technology, Cambridge.

Basjaruddin, NC., Kuspriyanto, Suhendar, Saefudin, D.,

Aziz, VA., 2016. Hardware Simulation of Automatic

Braking System Based on Fuzzy Logic Control.

Journal of Mechatronics, Electrical Power, and

Vehicular Technology. Vol.07, pp.1-6.

Pramudijanto, J., Ashfahani, A., Lukito, R., 2018,

Designing Neuro-Fuzzy Controller For

Electromagnetic Anti-Lock Braking System (ABS) On

Electric Vehicle, Journal of Physics.

Maghfiroh, H. et.al., 2016, Speed Control of a Single

Taipei Mass Rapid Transit System Train by Using a

Single Input Fuzzy Logic Controller, International

Journal of Electrical and Computer Engineering

(IJECE) Vol.6 No.2, pp.621-62

Design of Ground Vehicle System Semi-autonomous Preceder Type for Straight Path and Circular Path using Fuzzy Logic Method

183