=Paper= {{Paper |id=Vol-2694/paper2 |storemode=property |title=Fuzzy system as a method of controlling LEGO Linefollower vehicle using C# programming language |pdfUrl=https://ceur-ws.org/Vol-2694/p2.pdf |volume=Vol-2694 |authors=Krzysztof Grzesica,Jakub Wadas |dblpUrl=https://dblp.org/rec/conf/system/GrzesicaW20 }} ==Fuzzy system as a method of controlling LEGO Linefollower vehicle using C# programming language== https://ceur-ws.org/Vol-2694/p2.pdf

Fuzzy system as a method of controlling LEGO
Linefollower vehicle using C# programming language
Krzysztof Grzesicaa , Jakub Wadasa
a Faculty of Applied Mathematics, Silesian University of Technology, Kaszubska 23, 44-100 Gliwice

Abstract
In this article we present our implementation of intelligent system developed for a LEGO robot. We have built a robot which
is using sensors to trace the line and follow it. The model of tracing is based on fuzzy rules, which have been done to follow
the shape of a black line on the white background. The system was implemented using C# language for LEGO Mindstorm
elements. Results of our experiments show that the robot is able to follow various shapes of black lines.

Keywords
LEGO robot, Line detection, Fuzzy rules

1. Introduction lect such libraries and optimize them for robotic con-
structions. Also some approaches are based on frame-
In many fields of industry and production various ro- works, as proposed in [5] that frameworks are also
botics are applied to help in situations where the work possible for other constructions based on arduino el-
may be dangerous for humans or there is required high ements, which provide similar possibilities of robotic
precision at work. To start the research in such field constructions. All these what we want to create de-
we have selected LEGO mindstorm elements. This rob- pends on our ability to develop the construction and
ot model is easy to develop just by using simple bricks than to implement the control software. An interest-
in many shapes from the set. The set also provides ing discussion was given by [6, 7].
a programmable electronics with a variety of sensors. The other aspect of robotic models is to select a pro-
On the other hand such approach is widely presented per decision support model. In many IoT constructions
in other literature, what gave us an inspiration to de- we can find various approaches. In [8] was proposed
velop our idea. how to combine neural networks with rules in a form
In [1] a model of LEGO blocks was used to work of soft set table. While in [9] the fuzzy rule system
with human gestures, where sensors were used to read was implemented with neural networks to work in a
gestures and therefore take actions in mindstorm ele- smart house environment. It is also very popular to
ments. In [2] was presented how to use such LEGO develop fuzzy rules working on images, as discussed
models to help in first contact with machine intelli- in [10]. Mechanical vehicle constructions also very
gence. Authors describe good practices when using often benefit from artificial intelligence, both at con-
LEGO Mindstorm as a platform for programming ar- struction and simulation level, as proposed in [11, 12,
tificial intelligence systems. As presented in [3] these 13, 14].
models also support creative thinking, especially when Our approach is using C# language to implement
we develop new algorithms that must be programmed control system based on fuzzy rules to first detect and
in a special way accepted by the LEGO Mindstorm than follow the line. The LEGO Mindstorm robot is
platform. There are various approaches to use pro- using two servo-motors as accelerators of wheels con-
gramming languages in developing such robots. trolled by our developed artificial intelligence to fol-
However the most necessary is tu use an optimal low the line. our experiment shows that our construc-
library, which will support all necessary function that tion is well defined and can follow even the complex
provide optimal connection and configuration between lines on the board.
elements of the robot. In [4] was discussed how to se-

SYSTEM 2020: Symposium for Young Scientists in Technology, 2. Model
Engineering and Mathematics, Online, May 20 2020
" krzygrz684@student.polsl.pl (K. Grzesica); In this section we will describe how we have developed
jakuwad985@student.polsl.pl (J. Wadas) the model both from programming side and thinking

© 2020 Copyright for this paper by its authors. Use permitted under Creative
model.
Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073 CEUR Workshop Proceedings (CEUR-WS.org)
Figure 1: Vehicle construction

2.1. Project assumptions
The aim of this project was to build an artificial in-
telligence system based on fuzzy logic, which would
steer the vehicle used in LEGO Line-follower compe-
tition. Based on readings, the system must decide how
to change the speed of each engine, so that:
Figure 2: The bottom of the vehicle
• The vehicle does not go off the road (A moment
when a black line is not between vehicle’s wheels
is considered going off the road) 100
𝑃𝑅 = (2)
• The vehicle goes as fast as possible 𝑊 𝐶𝑅 − 𝐵𝐶𝑅
Where 𝑊 𝐶𝐿 and 𝐵𝐶𝐿 are accordingly, maximum and
The robot is connected to computer via Bluetooth. minimum reading value of left sensor. 𝑊 𝐶𝑅 and 𝐵𝐶𝑅
The robot sends raw readings and data to computer are maximum and minimum reading values of right
where all necessary calculations are being performed. sensor. These values are obtained during setup con-
Then, based on calculation results appropriate com- figuration. Reading output is determined using the fol-
mands from the computer are sent to robot. lowing formula:

2.2. Robot Construction 𝑟𝑒𝑎𝑑𝐿 = |(𝑟𝑒𝑎𝑑𝑖𝑛𝑔𝐿𝑒𝑓 𝑡 − 𝐵𝐶𝐿) ∗ 𝑃𝐿 − 100| (3)

To accomplish that task we have built a vehicle using 𝑟𝑒𝑎𝑑𝑅 = |(𝑟𝑒𝑎𝑑𝑖𝑛𝑔𝑅𝑖𝑔ℎ𝑡 − 𝐵𝐶𝑅) ∗ 𝑃𝑅 − 100| (4)
LEGO parts. The robot is based on intelligent EV3
Where 𝑟𝑒𝑎𝑑𝑖𝑛𝑔𝐿𝑒𝑓 𝑡 and 𝑟𝑒𝑎𝑑𝑖𝑛𝑔𝑅𝑖𝑔ℎ𝑡 are respectively
brick, it is equipped with two color sensors, which
raw vuales returned by left and right color sensor.
measure the level of reflected light they emit. The ve-
𝑟𝑒𝑎𝑑𝐿 and 𝑟𝑒𝑎𝑑𝑅 take values from 0 to 100, where 0
hicle is driven by two independent engines which di-
means total white, 100 - pure black.
rectly spin the wheels. The actual look of vehicle can
After that, the resultant reading value meaning the
be seen in Fig. 1 and Fig. 2.
inclination of the road is calculated accordingly to the
following formula:
3. Mathematical Model 𝑟𝑒𝑎𝑑𝑅 − 𝑟𝑒𝑎𝑑𝐿
𝑟𝑒𝑎𝑑𝑖𝑛𝑔 = (5)
3.1. Normalization 10

Firstly, we must normalize the readings from color sen- If both color sensors went off the road the variable
sors to negate the hardware differences. To do that we 𝑟𝑒𝑎𝑑𝑖𝑛𝑔 i set to 10 or -10 depending on the side of road
must first calculate parameters 𝑃𝐿 and 𝑃𝑅 accordingly the sensors are located.
as:
100
𝑃𝐿 = (1)
𝑊 𝐶𝐿 − 𝐵𝐶𝐿

10
Figure 3: Reading membership functions Figure 5: Application window before configuration

Figure 6: Application window after configuration
Figure 4: Speed membership functions

Table 1
3.2. Fuzzyfication Speed base of rules
To determine grade of membership Bell shaped and
Linear functions are used. The formulas are as follows: Left/Right rear slow medium fast
{ rear rear slow slow slow
1
1+| 𝑥−𝑐
,𝑦 ⩽𝑥 ⩽𝑧 slow slow slow medium medium
(6)
2𝑏
𝐵𝑒𝑙𝑙 = 𝑎 |
medium slow medium medium fast
0 ,𝑥 <𝑦∨𝑥 >𝑧
fast slow medium fast fast
{
1 ,𝑥 <0
𝐿𝑖𝑛𝑒𝑎𝑟 = (7)
0 ,𝑥 ⩾0
Where 𝑎, 𝑏 are parameters. Variables 𝑦 and 𝑧 are end- The results are being grouped by the rules they cor-
points of interval. Variable 𝑐 stands for center of func- respond with and for each of rule we choose the biggest
tion. In our project 𝑏 = 2.5 and 𝑎 = 2.5 in all but value. Lets assume that the biggest value for rule slow
one membership function. The membership function is 𝑢𝑤 , for medium - 𝑢𝑚 and for fast - 𝑢𝑓 . The final
Straight uses 𝑎 = 1.5. The reading membership func- resultant speed is obtained by using center of gravity
tions can be seen in Fig.3, while speed membership method defined as:
functions in Fig. 4. 𝑢𝑤 ∗ 𝑝𝑠 + 𝑢𝑚 ∗ 𝑝𝑚 + 𝑢𝑓 ∗ 𝑝𝑓
Then, the resultant vehicle speed composed from 𝑠𝑝𝑒𝑒𝑑 = (8)
𝑢𝑤 + 𝑢𝑚 + 𝑢𝑓
separate engines speeds is being calculated. In order to
determine it we calculate each engine’s grade of mem- where 𝑝𝑠 = 10, 𝑝𝑚 = 50 and 𝑝𝑓 = 90 are coefficients of
bership for all of speed membership functions and ac- speed rules.
cordingly to fuzzy rules base shown in Tab. 1 we con-
duct necessary calculations.

11
Table 2
Inclination base of rules

Speed/Reading HL EL S ER HR
fast HL HL S HR HR
medium HL EL S ER HR
slow HL EL S ER HR
HL - Hard Left, EL - Easy Left, S - Straight,
ER - Easy Right, HR - Hard Right
Figure 7: Reading method

3.3. Deffuzyfication 𝑖𝑛𝑐
𝑅𝑖𝑔ℎ𝑡𝑀𝑜𝑡𝑜𝑟 = (19)
The next step is to calculate the inclination, which con- 4
sists of resultant speed and resultant reading values. where 𝑖𝑛𝑐 is the inclination value calculated before.
In order to determine it we calculate resultant speed In case of Easy Left and Easy Right if the 𝑖𝑛𝑐 value is
grade of membership for all of speed membership func- lesser than -50 or greater than 50, then it is set to 50.
tions and resultant reading grade of membership for
all of reading membership functions. Then, accord-
ingly to fuzzy rules base shown in Tab. 2 we conduct 4. Implementation
necessary calculations. The results are being grouped
by the rules they correspond with and for each of rule Let us now present the software we have done.
we choose the biggest value. Lets assume that the big-
gest value for rule HL is 𝑢1 , for EL - 𝑢2 , for S - 𝑢3 , for ER 4.1. Application
- 𝑢4 and for HR - 𝑢5 . The final inclination is obtained
Based on the mathematical model described above, we
by using center of gravity method defined as:
have created a robot control application. The applica-
tion has a very simple graphical user interface, which
𝑖𝑛𝑐 =
𝑢1 ∗ 𝑝1 + 𝑢2 ∗ 𝑝2 + 𝑢3 ∗ 𝑝3 + 𝑢4 ∗ 𝑝4 + 𝑢5 ∗ 𝑝5 makes it user-friendly. In Fig. 5 and Fig. 6 application
𝑢1 + 𝑢2 + 𝑢3 + 𝑢4 + 𝑢5 window can be seen.
(9)
where 𝑝1 = −100, 𝑝2 = −50, 𝑝3 = 0, 𝑝4 = 50, 𝑝5 = 100
are coefficients of inclination rules accordingly for HL, 4.2. Program code
EL, S, ER and HR. The program is written in 𝐶#. In addition to the stan-
Finally, the rule with the greatest inclination grade dard .𝑁 𝐸𝑇 libraries, the program uses the 𝐿𝑒𝑔𝑜.𝐸𝑣3
of membership value is chosen. Depending on it, the library for communication between the computer and
engines speeds are calculated and set. For Hard Left the robot. The program has been divided into classes
(10) and (11), for Easy Left (12) and (13), for Straight and appropriate methods.
(14) and (15), for Easy Right (16) and (17) and for Hard
Right (18) and (19). Reading method This method is responsible for the
𝑖𝑛𝑐 normalization of color sensor data. This method also
𝐿𝑒𝑓 𝑡𝑀𝑜𝑡𝑜𝑟 = (10) serve as a precaution against loosing the route by a
4
robot. Method code can be seen in Fig. 7.
𝑅𝑖𝑔ℎ𝑡𝑀𝑜𝑡𝑜𝑟 = 𝑖𝑛𝑐 (11)
𝐿𝑒𝑓 𝑡𝑀𝑜𝑡𝑜𝑟 = 𝑖𝑛𝑐 ∗ 1.5 (12) HowTheRouteRuns method This method is a frag-
𝑅𝑖𝑔ℎ𝑡𝑀𝑜𝑡𝑜𝑟 = 𝑖𝑛𝑐 ∗ 2 (13) ment of the fuzzy system. It combines the route incli-
nation with the current vehicle speed and decides how
𝐿𝑒𝑓 𝑡𝑀𝑜𝑡𝑜𝑟 = 100 (14) to react based on that data. A piece of the method code
𝑅𝑖𝑔ℎ𝑡𝑀𝑜𝑡𝑜𝑟 = 100 (15) is in Fig. 8.
𝐿𝑒𝑓 𝑡𝑀𝑜𝑡𝑜𝑟 = 𝑖𝑛𝑐 ∗ 2 (16)
Robot class This is the class which object repre-
𝑅𝑖𝑔ℎ𝑡𝑀𝑜𝑡𝑜𝑟 = 𝑖𝑛𝑐 ∗ 1.5 (17) sents vehicle instances in the program. The most im-
portant class fields and properties representing the state
𝐿𝑒𝑓 𝑡𝑀𝑜𝑡𝑜𝑟 = 𝑖𝑛𝑐 (18)

12
Figure 11: ChooseTurn method

Figure 8: HowTheRouteRuns method

Figure 12: TurnHard method

Figure 9: Robot class

Figure 10: Go method

of the object can be seen in Fig. 9. 𝐵𝑟𝑖𝑐𝑘 class ob-
ject, which belongs to the 𝐿𝑒𝑔𝑜.𝐸𝑣3 library represents Figure 13: Test route
the LEGO EV3 brick. 𝐶𝑜𝑛𝑛𝑒𝑐𝑡𝑒𝑑 property represents
the state of connection between the program and the
robot. Fields 𝑙𝑒𝑓 𝑡𝑆𝑒𝑛𝑠𝑜𝑟 and 𝑟𝑖𝑔ℎ𝑡𝑆𝑒𝑛𝑠𝑜𝑟 store values
The method is responsible for choosing the direction
obtained from color sensors. Fields _𝑙𝑒𝑓 𝑡𝑀𝑜𝑡𝑜𝑟 and
in which the robot should go. The code for this method
_𝑟𝑖𝑔ℎ𝑡𝑀𝑜𝑡𝑜𝑟 store current motors speeds.
is shown in Fig. 11.

Go method This method belongs to the 𝑅𝑜𝑏𝑜𝑡 class.
TurnHard method This method also belongs to the
The method is responsible for the robot’s movement.
𝑅𝑜𝑏𝑜𝑡 class. It is responsible for updating the variables
It contains loop in which the fuzzy system makes cal-
representing the current motors power. The full code
culations, decides about the next move and finally ma-
of the method can be seen in Fig. 12. The 𝑅𝑜𝑏𝑜𝑡 class
kes that move by setting the power of motors. The full
also has 𝑇𝑢𝑟𝑛𝐸𝑎𝑠𝑦 and 𝐺𝑜𝑆𝑡𝑟𝑎𝑖𝑔ℎ𝑡 methods whose
code of the method can be found in Fig. 10.
operations are analogous to the 𝑇𝑢𝑟𝑛𝐻 𝑎𝑟𝑑 method.
ChooseTurn method This method belongs to the
𝑅𝑜𝑏𝑜𝑡 class. It is called in the loop described above.

13
Figure 14: A part of a test ride

5. Tests Interactive Learning Environments 27 (2019)
293–306.
In order to test how the vehicle performs we created a [3] N. L. Fanchamps, L. Slangen, P. Hennissen,
test route (Fig. 13). The route is a 19-millimeter-wide M. Specht, The influence of sra programming
black line on a white background. It is quite compli- on algorithmic thinking and self-efficacy using
cated due to a lot of 90◦ angle turns, it does not contain lego robotics in two types of instruction, Inter-
crossroads though. The vehicle runs through the route national Journal of Technology and Design Edu-
precisely but with little speed. It is worth mentioning cation (2019) 1–20.
that the system is incredibly sensitive to any changes [4] A. Spanò, A. Cortesi, Legodroid: A type-driven
in lightning and the non-uniformity of it. A part of a library for android and lego mindstorms interop-
test ride can be seen in Fig. 14. erability, Sensors 20 (2020) 1926.
[5] J. Vega, J. M. Cañas, Pybokids: An innova-
tive python-based educational framework using
6. Conclusion real and simulated arduino robots, Electronics 8
Because of poor functioning of some of 𝐿𝑒𝑔𝑜.𝐸𝑉 3 li- (2019) 899.
brary functions that our project was based on, the re- [6] J. Wang, X. Du, H. Wang, Research & imple-
sults were not as good as we had imagined. Commands mentation of multitasking lego robots, in: 2019
must have been initialized for a specified time amount IEEE 4th International Conference on Advanced
which led to delays and unstable movement of vehicle. Robotics and Mechatronics (ICARM), IEEE, 2019,
What is more, in order to work properly the program pp. 655–659.
had to use function 𝐷𝑒𝑙𝑎𝑦(), which stops the robot [7] R. Giuliano, G. Cardarilli, C. Cesarini, L. Di Nun-
suddenly for few milliseconds. zio, F. Fallucchi, R. Fazzolari, F. Mazzenga, M. Re,
Considering that 𝐿𝑒𝑔𝑜.𝐸𝑉 3 library has not been up- A. Vizzarri, Indoor localization system based on
dated for 7 years and it’s author officially abandoned bluetooth low energy for museum applications,
the project, rewriting the whole project to 𝑅𝑜𝑏𝑜𝑡𝐶 lan- Electronics (Switzerland) 9 (2020) 1–20.
guage seems to be the best way to improve program’s [8] M. Woźniak, D. Połap, Soft trees with neural
performance, allowing the program to run on EV3 brick components as image-processing technique for
itself. Apart from solving problems mentioned above, archeological excavations, Personal and Ubiqui-
that approach would also terminate problems connect- tous Computing (2020) 1–13.
ed with Bluetooth connection latency. [9] M. Woźniak, D. Połap, Intelligent home systems
for ubiquitous user support by using neural net-
works and rule based approach, IEEE Transac-
References tions on Industrial Informatics (2019).
[10] G. Capizzi, G. Lo Sciuto, C. Napoli, D. Polap,
[1] L. I. Kovács, Gesture-driven lego robots, Acta M. Woźniak, Small lung nodules detection based
Universitatis Sapientiae, Informatica 11 (2019) on fuzzy-logic and probabilistic neural network
80–94. with bio-inspired reinforcement learning, IEEE
[2] Á. Martínez-Tenor, A. Cruz-Martín, J.-A. Transactions on Fuzzy Systems 6 (2020).
Fernández-Madrigal, Teaching machine learning [11] J. T. Starczewski, P. Goetzen, C. Napoli, Triangu-
in robotics interactively: the case of rein- lar fuzzy-rough set based fuzzification of fuzzy
forcement learning with lego® mindstorms,

14
rule-based systems, Journal of Artificial Intel-
ligence and Soft Computing Research 10 (2020)
271–285.
[12] M. Woźniak, D. Połap, Hybrid neuro-heuristic
methodology for simulation and control of dy-
namic systems over time interval, Neural Net-
works 93 (2017) 45–56.
[13] G. Capizzi, F. Bonanno, C. Napoli, Hybrid neural
networks architectures for soc and voltage pre-
diction of new generation batteries storage, 2011,
pp. 341–344.
[14] F. Bonanno, G. Capizzi, C. Napoli, Some remarks
on the application of rnn and prnn for the charge-
discharge simulation of advanced lithium-ions
battery energy storage, 2012, pp. 941–945.