Applications of fuzzy logic in the control of robotic manipulators

Fuzzy logic has been utilized at several hierarchical levels of a typical robotic control system. Four broad levels of application may be identified - task design, system monitoring (including self-tuning and self-organization), information filtering and preprocessing, and in-loop direct control. Even though the need for fuzzy logic is felt mostly at upper levels of the control system, the present applications are mainly concentrated within the lowest level, perhaps driven by convenience rather than necessity. This paper surveys several applications of fuzzy logic in the control of robotic manipulators. Applications are grouped into four hierarchical categories, broadly corresponding to an existing architecture of a robotic control system. Such a classification can be beneficial in ascertaining the appropriateness of fuzzy logic for the specific control task.