Disturbance Observer-Based Adaptive Finite-Time Singular Perturbation Constrained Control for Flexible Joint Manipulators

This paper proposes a disturbance observer-based adaptive finite-time singular perturbation control scheme for flexible joint manipulators with state constraints. Firstly, a fuzzy logic system-based observer is designed to estimate unknown external disturbances. Then, a fuzzy adaptive finite-time singular perturbation controller is developed to address model uncertainties and improve the response speed of the rigid subsystem. In particular, the singular perturbation method avoids the design of unnecessary virtual control laws and error compensation signals by decoupling the original system into the reduced-order rigid and fast subsystems, which reduces the computational burden. Stability analysis verifies that the closed-loop signals converge within finite time, while ensuring that all states of the rigid subsystem remain within constraint bounds. Finally, the effectiveness of the proposed control scheme is demonstrated by simulation results.