Finite-Time Model Reference Adaptive Grasping Control With Fuzzy State Observer for Maglev Grasping Robot System

Abstract

The Maglev grasping robot system (MGRS) with single axial grasping coils is proposed to achieve the reliable transporting, but the grasped body is always subjected to multiple disturbances derived from the robot motion. Hence, the finite-time model reference adaptive control is presented to guarantee the satisfied transient and robustness performance. First, two degree-of-freedom dynamic model is built with consideration of uneven magnetic flux. Second, the prescribed performance reference model (PPRM) is designed by finite-time control (FTC), and its convergence performance is analyzed by FTC Lyapunov function. Then, the adaptive tracking controller is deduced from the model matching and FTC stability, which makes the MGRS to approximate the PPRM in finite time, and fuzzy approximation state observer is used to online get external disturbances and unknown state. It is proved that the proposed strategy can guarantee the closed system semiglobally practical finite-time stability, and the model approximation error and observe error can converge to the desired neighborhood in finite time. Finally, experimental results show that the MGRS with the proposed strategy has faster transient performance, and stronger robustness than the other three strategies.