Robot manipulator task control with obstacle avoidance using fuzzy behavior-based strategy

Palitha Dassanayake, Keigo Watanabe, Kazuo Kiguchi, Kiyotaka Izumi

Research output: Contribution to journalArticlepeer-review

5 Citations (Scopus)


In this paper, the concept of fuzzy behavior-based control is used to construct a fuzzy generator that generates the desired positions and orientations of a robot manipulator in the Cartesian space. A servo controller is introduced between the fuzzy trajectory generator and the robot. This method is proposed to minimize the drawbacks in extending a fuzzy behavior-based control used previously, while keeping the advantages of the fuzzy behavior-based strategy. For the PUMA robot, the direct extended version of the control system applied to a three-link manipulator in a previous work, is compared with the proposed method. Two methods are first applied for two behavior groups without any obstacle, in which fuzzy behavioral elements in each method are trained by a genetic algorithm. It is proved that a desired result is not possible within a few numbers of generations for the extended method, whereas the proposed method is able to achieve good results. Moreover, the proposed method is simulated to prove the benefit of the method for three behavior groups with an obstacle. Therefore, it can be concluded that the present approach is suitable in task control of high degree-of-freedom multi-link manipulators while avoiding obstacles for manipulators similar to PUMA robot.

Original languageEnglish
Pages (from-to)139-158
Number of pages20
JournalJournal of Intelligent and Fuzzy Systems
Issue number3-4
Publication statusPublished - Dec 1 2001
Externally publishedYes


  • Behavior-based control system
  • Fuzzy control
  • Genetic algorithms
  • Obstacle avoidance
  • PUMA robot manipulator

ASJC Scopus subject areas

  • Statistics and Probability
  • General Engineering
  • Artificial Intelligence


Dive into the research topics of 'Robot manipulator task control with obstacle avoidance using fuzzy behavior-based strategy'. Together they form a unique fingerprint.

Cite this