Learning of fuzzy-behaviours using Particle Swarm Optimisation in behaviour-based mobile robot
by Andi Adriansyah, Shamsudin H.M. Amin
International Journal of Intelligent Systems Technologies and Applications (IJISTA), Vol. 5, No. 1/2, 2008

Abstract: Behaviour-based mobile robots should have an ideal controller to generate perfect behaviour action. A schema to overcome these problems is provided, known as Fuzzy Behaviour-based robot. However, tuning fuzzy parameters is not a simple effort. This paper presents a technique to tune automatically fuzzy Rule Bases and fuzzy Membership Functions (MF) by Particle Swarm Optimisation (PSO), named as Particle Swarm Fuzzy Controller (PSFC). The behaviours are controlled by PSFC to generate individual command action. Later, a Context-Dependent Blending (CDB) based on meta-fuzzy rules coordinates the commands to produce final control action. A Sigmoid Decreasing Inertia Weight has been proposed for a good balancing of global and local searches for obtaining good convergence speed and best accuracy of PSO algorithm. The algorithm is validated using parameters of MagellanPro mobile robot and tested by simulation using MATLAB/SIMULINK. Simulation results show that the proposed model offers hopeful advantages and has improved performance.

Online publication date: Mon, 05-May-2008

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Intelligent Systems Technologies and Applications (IJISTA):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com