Abstract- This paper focuses on the Genetic Algorithm
learning paradigm applied to train the ANNs for balancing
the cart-pole balancing system. The studied system is a
classic control problem namely “cart-pole” problem. We will
apply the unconventional techniques Artificial Neural
Network, Genetic Algorithm and Fuzzy Logic to a classic
control problem "cart-pole”. In this paper we have tried to
train the Artificial Neural Network (ANN) with using
Genetic Algorithms (program is written in MATLAB) which
is compared with the output obtained using the Artificial
Neural Network Toolbox provided in MATLAB.
In proposed approach we have used both ANNs and
Genetic Algorithm to get more optimal solution. Here we
applied the approach for the Fuzzy logic technique to design
a Fuzzy Logic Controllers (FLC) using ANNs and Genetic
Algorithm (GA). The Fuzzy rules which are needed to
control the problem will be framed with the combination of
Artificial Neural Networks and Genetic Algorithm. It has
been found that such a searching technique converges
intelligently and much faster than conventional learning
means. Performance of the presented neural network
training using the genetic algorihtms is much better and
providing more accurate results.
Keywords: Artificial Neural Network, Genetic Algorithm,
Fuzzy Logic Controllers.
I. INTRODUCTION
The control of a cart-pole system is widely used as a
benchmark problem for testing the efficiency of
reinforcement learning algorithms [8]. Artificial Neural
Networks can be trained to simulate the execution of the
rule base of the Fuzzy Logic Controllers (FLC) using
Genetic Algorithms(GA) for determining the solution of
the cart-pole problem (program is written in
MATLAB).The genetic design approach discussed in this
paper offers a convenient and complete way to design a
fuzzy controller in the shortest time. We are expecting
better solution in terms of number of iterations,
performance of the presented neural network training
using the genetic algorihtms must be much better with
more accurate results. Again in terms of requirement of
fewer epochs for training the ANN using GA comparing
to ANN toolbox provided in MATLAB.
We will describe two earlier applications of genetic
algorithms to the automatic generation of a fuzzy rule
base, and compare these with the ANN toolbox output to
our approach. One of the earliest applications of genetic
algorithms to the design of fuzzy systems was developed
by Karr (1991), again to solve the cart-pole problem [2].
The production of the fuzzy controller begins with the
definition of the fuzzy sets used to describe each input
variable. Each of the four input variables are characterized
by three fuzzy sets—NEGATIVE, ZERO, and
POSITIVE—yielding 81 possible combinations. The
fuzzy system designer then assigns one of seven choices
for the output to each input combination. The resulting
fuzzy system represents the expert's "best guess". The
membership function extrema are then encoded into a bit
string, and a genetic algorithm is applied to shift the
membership functions so as to find locations which
improve performance. The evolved system consistently
outperforms the original, being capable of recovering
from initial positions that fail under the original rule base.
In second application, Genetic algorithms for
automatic design of fuzzy logic controllers have been
developed [9], using sophisticated membership functions
that intrinsically reflect the nonlinearity encountered in
many engineering applications. But the sophistication
obtained by the machine based automatic design could not
be reached by manual design which is exclusively based
on a painstaking trial-and-error process. As the number of
variables increases the length of the string encoding the
system size increases exponentially, with a corresponding
exponential increase in the complexity of the search space.
Such a system is unlikely to scale well for more complex
problems.
We are using ANNs and Genetic Algorithm both to
get more optimal solution. Here we applied the approach
for the fuzzy logic technique to design a Fuzzy Logic
Controller (FLC) using ANNs and Genetic Algorithm
(GA). The resultant optimal fuzzy logic controller is used
in centering a cart. Artificial Neural Networks can be
trained to simulate the execution of the rule base of the
Fuzzy Logic Controllers (FLC) using Genetic Algorithms.
Also the training of artificial neural networks using
genetic algorithms is extended to include a priori control
knowledge of human operators in the form of rule base
table. It is shown that the system can solve more
concretely a fairly difficult control learning problem. It
also demonstrated the feasibility of the method when
applied to a cart-pole balancing problem. The
performance of the GA and ANN Optimized Fuzzy Logic
controller is compared with that of the conventional ANN
EVOLUTIONARY DESIGN OF FUZZY LOGIC CONTROLLERS WITH THE
TECHNIQUES ARTIFICIAL NEURAL NETWORK AND GENETIC
ALGORITHM FOR CART-POLE PROBLEM
Reena Thakur
1
, Vinay Kr Singh
2
, Manu Pratap Singh
3
1
Department of Information Technology, AEC, Agra, India
2
Department of Information Technology, AEC, Agra, India
3
Department of Computer Science, ICIS, Dr. B.R. Ambedkar University, Khnadari Campus, Agra, India
(
1
rina151174@rediffmail.com,
2
vksingh100@rediffmail.com,
3
manu_p_singh@hotmail.com)
978-1-4244-5967-4/10/$26.00 ©2010 IEEE