fuzzy logic

Topics: Machine learning, Robotics, Fuzzy logic Pages: 15 (2516 words) Published: February 7, 2014
EMBEDDED LEARNING ROBOT WITH FUZZY Q-LEARNING
FOR OBSTACLE AVOIDANCE BEHAVIOR
Khairul Anam1,Prihastono2,4,Handy Wicaksono3,4, Rusdhianto Effendi4, Indra Adji S5, Son Kuswadi5, Achmad Jazidie4, Mitsuji Sampei6
1

Department of Electrical Engineering, University of Jember, Jember, Indonesia (Tel : +62-0331-484977 ; E-mail: kh.anam.sk@gmail.com)
2
Department of Electrical Engineering, University of Bhayangkara, Surabaya, Indonesia (Tel : + 62-031-8285602; E-mail: prihtn@yahoo.com)
3
Department of Electrical Engineering, Petra Christian University, Surabaya, Indonesia (Tel : +62-031-8439040; E-mail: handy@petra.ac.id )
4
Department of Electrical Engineering, Institut Teknologi Sepuluh Nopember, Surabaya, Indonesia (Tel : +62 031-599 4251; E-mail: ditto@ee.its.ac.id, jazidie@ee.its.ac.id) 5
Electronics Eng. Polytechnics Institute of Surabaya, Surabaya Indonesia (Tel : +62 031-5947280; E-mail: indra@eepis-its.edu , sonk@eepis-its.edu) 6
Department of Mechanical and Control Engineering, Tokyo Institute of Technology, Tokyo, Japan (Tel : +81-3-5734-2552; E-mail: sampei@ctrl.titech.ac.jp)
Abstract: Fuzzy Q-learning is extending of Q-learning algorithm that uses fuzzy inference system to enable Qlearning holding continuous action and state. This learning has been implemented in various robot learning application like obstacle avoidance and target searching. However, most of them have not been realized in embedded robot. This paper presents implementation of fuzzy Q-learning for obstacle avoidance navigation in embedded mobile robot. The experimental result demonstrates that fuzzy Q-learning enables robot to be able to learn the right policy i.e. to avoid obstacle.

Keywords: fuzzy q-learning, obstacle avoidance

EMBEDDED LEARNING ROBOT USING FUZZY Q-LEARNING FOR
OBSTACLE AVOIDANCE BEHAVIOR

ABSTRACT
Fuzzy Q-learning is extending of Q-learning
algorithm that uses fuzzy inference system to enable Qlearning holding continuous action and state. This learning has been implemented in various robot
learning application like obstacle avoidance and
target searching. However, most of them have not been
realized in embedded robot. This paper presents
implementation of fuzzy Q-learning for obstacle
avoidance navigation in embedded mobile robot. The
experimental result demonstrates that fuzzy Q-learning
enables robot to be able to learn the right policy i.e. to
avoid obstacle.
Keywords : behavior based control, fuzzy q-learning
1. Introduction
In unstructured environment, a big change may be
happen suddenly. To overcome it, robot control must
be able to change its control action to adapt with new
condition. Therefore, it is needed control system for
robot that can learn its environment.
Because the environment is unstructured and
unknown, unsupervised learning is suitable used for
enabling the robot to learn its environment. For this
purpose, reinforcement learning methods have been
receiving increased attention for use in autonomous
robot systems. Reinforcement learning can be realized
using Q-learning. However, since Q-learning deals
with discrete actions and states, an enormous amount
of states may be necessary for an autonomous robot to
learn an appropriate action in a continuous
environment. Therefore, Q-learning can not be directly
used to such a case due to the problems of the curse of
dimensionality.
To overcome this problem, variations of the Qlearning algorithm have been developed. Different authors have proposed to use the generalization of
statistical method (hamming distance ,statistical
clustering)[3], of generalization ability of feed-forward
Neural Networks to store the Q-values[3,5,9]. Another
approach consist in extending Learning into fuzzy
environments [4] and was called by fuzzy q-learning.
In this approach, prior knowledge can be embedded
into the fuzzy rules which can reduce training
significantly. Therefore, this approach is used in this
paper.
Fuzzy Q-learning...
Continue Reading

Please join StudyMode to read the full document

You May Also Find These Documents Helpful

  • Design of Fuzzy Controller for Two Tank Interacting System Essay
  • Fuzzy Logic Essay
  • Essay about Fuzzy Logic Controller
  • Precise Fuzzy Inference Systems Essay
  • Field Oriented Control of PMSM with Model Reference Adaptive Control Using Fuzzy-PI Controller Essay
  • A High Performance Induction Motor Drive System Using Fuzzy Logic Controller Essay
  • Essay about Fuzzy Controlled Architecture for Perfomance Tuning of Dbms
  • Nfis: Adaptive Neuro-Fuzzy Inference Research Paper

Become a StudyMode Member

Sign Up - It's Free