Approximating a robot inverse kinematics solution using fuzzy logic tuned by genetic algorithms

C. Y. Chen, M. G. Her, Y. C. Hung, Mansour Karkoub

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

A new scheme based on recursive fuzzy logic is presented in this paper for solving the point-to-point inverse kinematics problem of serial robots. To improve the convergence problem in the whole workspace, the membership functions of the fuzzy logic are searched for, tuned, and optimised using a simple genetic algorithm. A dominant joint, which brings the end-effect closer to the desired target, has to be selected before the implementation of the fuzzy logic in order to reduce the number of fuzzy logic iterations. The inverse kinematics solution of robots is usually obtained by direct inversion of the kinematics equations, but this technique often leads to a singular Jacobian matrix during the calculations. The work presented in this paper provides a direct approach to the calculation of the kinematics inverse problem which bypasses the kinematic singularities. Computer simulations of the proposed scheme confirm the findings of the theoretical developments.

Original languageEnglish
Pages (from-to)375-380
Number of pages6
JournalInternational Journal of Advanced Manufacturing Technology
Volume20
Issue number5
DOIs
Publication statusPublished - 2002
Externally publishedYes

    Fingerprint

Keywords

  • Fuzzy logic
  • Genetic algorithm
  • Inverse kinematics

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Mechanical Engineering
  • Computer Science Applications
  • Industrial and Manufacturing Engineering

Cite this