Advanced fuzzy observer-based fault identification for robot manipulators

Abstract

Designing an effective procedure for fault detection and identification (FDI) is necessary to maintain the healthy and safe operation of robot manipulators. The complexities of nonlinear parameters inherent in a robot manipulator make it challenging to detect and identify faults. To address this issue, a powerful, robust, hybrid fault identification method based on the fuzzy extended ARX-Laguerre proportional integral (PI) observer for perturbation robot manipulators is presented. Accurate fault estimation is an essential challenge in classical extended ARX-Laguerre PI observers. The Takagi-Sugeno (T-S) fuzzy algorithm is applied to the sliding mode extended ARX-Laguerre PI observer to modify the performance of fault estimation. Moreover, using the ARX-Laguerre algorithm, PI observation technique, sliding mode estimation method, and T-S fuzzy procedure, the system’s performance showed fast convergence and high accuracy. A PUMA robot manipulator was used to test the effectiveness of the proposed method. Results indicated that the proposed algorithm outperforms the ARX-Laguerre PI observer performance.