Académique Documents
Professionnel Documents
Culture Documents
STSIVA 2008
Abstract—There are different forms to design adaptive rotation it selves to the sense objects around”. Then the
fuzzy controllers applied to robotic systems, this work adaptive controller is therefore “intuitively” a controller that
changes the traditional methods like the common “Fuzzy can modify its movement behaviour according the
Model Reference Learning Controller” FMRLC, see[1] - environment (number of objects and their distance).
[2], which is supported in change of rule base to modify
the controller parameters. Here the purpose is to maintain FUZZYBOT is assembled by two ultrasonic sensors, these
constant the rule base and change fuzzy sets size to control are located on the frontal panel, it has two servo motors
a robot tricycle named FUZZYBOT, also presents figures reformed to do left and right completely rotations, this robot
with five fuzzy sets inputs and four fuzzy sets outputs, use a emission circuit and a reception circuit supported by a
universe to these sets are modify to produce others fuzzy micro controlled target and DSP evaluation target [3].
sets according to objects around the robot; on the other
hand, motors’ robot also change its rotation to avoid crash Devices able to run speedily programs have been
always controlled by fuzzy logic system. There are shown developed, these are “the Digital Signal Processors” DSPs, its
statistical probes in order to compare numerical examples main characteristics are: micro size and processing millions
of how the performance of the controller purposed on instructions per second. It has many resources, for example:
relate to others conventional controllers in real systems serials communications, parallel ports, PWM, timers.
using this robot. Associate DSPs with intelligence artificial algorithms has
many advantages; just as run freely these programs without
Terms—Adaptive fuzzy control, fuzzy sets, intelligent computers or stronger processors, and then, these systems
control. could be probed easily on robotics applications.
I. INTRODUCTION
II. ADAPTIVE FUZZY LOGIC SYSTEM
The robotics in this moment is an important area for
different fields of science; there are many applications using Known the characteristics of FUZZYBOT, the main
robots, especially at the factories, on electro medicine and objective of this work is to achieve a free movement robot on
home services. The programs on these classes of devices are a plain area with objects inside, motors’ robot have to control
also important too. Artificial intelligence allows to develop this device through DSP orders avoid crashes, for this goal,
efficiently algorithms for robotics, for example: fuzzy logic, we have to design a AFLS “Adaptive Fuzzy Logic System”
neuronal links and genetics algorithms. But currently many algorithm. This objective permits us to deduce that the objects
application uses on theirs controller’ programs fuzzy logic distance is the most important variable like the motors
algorithms; this theory has been proved over different movements, then on our AFLS, distance is the input variable
applications, especially when the control is no linear, see [1]- and motors’ movement is the output variable.
[2]. Fuzzy logic can be described as computing with words
rather than numbers, and fuzzy control can be described as
control with sentences. It is more natural to use sentences, or
rules, for instance operator controlled plants, with the control
strategy written in terms of if-then clauses. This work presents
a new fuzzy control strategy without human intervention, this
is an adaptive fuzzy control applied in a robotic system named
FUZZYBOT. Then, it has an adaptive behaviour, this
characteristic is supported over the changes of its algorithms,
it
means to modify the program according to changing some Fig. 1. fuzzy model reference learning controller
environments circumstances, for example: “adapted its motor
XIII SIMPOSIO DE TRATAMIENTO DE SEÑALES, IMÁGENES Y VISIÓN ARTIFICIAL. STSIVA 2008
Classical fuzzy controller parameters are changed to [19] AYANNA, Howard, & HOMAY-OUN Seraji, & TUNSTE Edward l. A
Rule-Based Fuzzy Traversability Index for Mobile Robot Navigation,
achieve adaptive characteristics on the fuzzy system, this IEEE International Conference on Robotics and Automation, Seoul
design is based on Velásquez, F. Torre, C. Muñoz y C. Shene Korea, Mayo 2001. http://citeseer.ist.psu.edu/howard01rulebased.html.
work, they change output sets size to control a plant; [20] KONOLIGE, K. Myers. The Saphira architecture for autonomous
mobile robots, pp. 211-242, 1998.
successful results obtained on this control permit us
[21] TIMOTHY, J. Ross.Fuzzy Logic with engineering applications.
continuing working with this class of adaptive fuzzy control, [22] MPLAB USER’S GUIDE, Microchip Technology, 2005.
see [10], [11].
REFERENCES
[1] KOO, John Tak Kuen. Design of Stable Adaptive Fuzzy Control.
University of California at Berkeley, Vol. 1, pp. 1-130, 1997.
[2] PAN Lei & WOO Peng Yung. “PD Manipulador controller with
Adaptive Compensation”. Vol 1. pp. 1-43, 1996.
[3] Procesadores de señales digitales 53f8323 EVM Rev 3.0, 1/30/2004.
[4] SUGENO, M. “industrial applications of fuzzy control” El servier
Science. Vol 1, pp. 10-115, 199.
[5] TOMEI, P. “Adaptive PD Controller for Robot Manipulator”, IEEE.
[6] VELÁSQUEZ, F. Torres, C. Muñoz y C. Shene, “Control Difuso
Adaptivo y su Aplicación a una Planta Piloto de Secado por Lecho
Fluidizado”, Revista Facultad de Ingeniería de la Universidad de
Tarapaca, Información Tecnológica-Vol. 16 N°5-2005.
[7] SAFOTT,I Alan. The uses of fuzzy logic in autonomous robot
navigation., Soft Computing Vol1, pp. 180-197, 1997.
[8] WANG Li Xing. “Stable Adaptive Fuzzy Control of Non linear
Systems” IEEE Transations on fuzzy Sistems Vol. 1, No 2, pp. 146-155
mayo 1993.
[9] LEWIS, Jagan nathan & A. Yesildirek, “Neural Network Control of
Robot Manipulators and Non linear Systems”, Vol 1, pp. 56-74,
Philadelphia 1999.
[10] CALDERÓN, Dora María. Control difuso adaptable en un
servomecanismo no lineal, Vol 1, México, Noviembre de 2003.
[11] DUARTE, Oscar Germán, UNFUZZY. UNAL,1998.
[12] AYANNA, Howard, & TUNSTEL, Edward & DEAN, Edwards &
CARLSON, Alan, Enhancing Fuzzy Robot Navigation Systems, Canada,
Julio 2001.
[13] DRIANKOV, D. Hellen doorn & REINFRANK, M., “An Introduction
to Fuzzy Control ”New York 1996.
[14] JANG, S. Anfis. Adaptive Network based Fuzzy Inference Systems”,
IEEE Transactions on Systems and Cybernetics, Vol. 23, No.3, pp.665-
685,Mayo1993.
[15] KLIR, J. George & FOLGER Tina A. Fuzzy Sets Uncertainty and
Information, Vol 1, pp. 22-54 New York 1992.
[16] DRIANKOV, D. & HELLENDORN, H. Fuzzy Logic with Unless –
Rules. Report IDA-RKL-92TR50
[17] Transaction on Automatic Control, Vol 7, No. 4pp. 565-570, 1991.
[18] AYANNA Howard, HOMAYOUN Seraji A. Real-Time Autonomous
Rover Navigation System , In Proc. of the World Automation Congress,
2000. http://robotics.jpl.nasa.gov/people/tunstel/papers/ifsa_nafips01.pdf
.