Adaptive fractional-order non-singular terminal sliding mode control based on fuzzy wavelet neural networks for omnidirectional mobile robot manipulator.

This paper studies a novel adaptive fractional-order non-singular terminal sliding mode (FO-NTSM) control strategy for omnidirectional mobile robot manipulator (OMRM) with unknown parameters and external disturbances. Firstly, we adopt the fuzzy wavelet neural networks (FWNNs) to estimate the dynamic uncertainty of the OMRM because it has superior function approximation capability. Secondly, we design the adaptive NTSM controller to attenuate external disturbances by virtue of adjusting the weights of the FWNNs online. Moreover, we obtain a fractional-order (FO) control criterion, which speed up the convergence of the algorithm. In addition, we prove the globally robust stability of the OMRM control system through a designed Lyapunov function. Finally, simulation and experiment researches indicate the feasible and validity of the presented method. • By applying fractional calculus, a novel adaptive sliding mode controller with FO is proposed to investigate the trajectory tracking for OMRM system in present of unknown parameters and external disturbances. • The superior combination of FWNNs and robust techniques not only largely overcomes the model uncertainty of the controlled object, but also is more robust to external unfavorable factors. • The designed controller is applied to a 3-DOF OMRM, accordingly. It is verified by experiments that the actual control performances are considerable. [ABSTRACT FROM AUTHOR]