An adaptive fuzzy finite-time singular perturbation control is proposed for flexible joint manipulators with state constraints. First, the flexible joint manipulator system is decoupled into a rigid subsystem and a fast subsystem through singular perturbation technique. Second, a finite-time controller is introduced to improve the response speed of the rigid subsystem so that it can converge within a finite time. And then, all the rigid subsystem states are confined within the scope of the constraint by the barrier Lyapunov function. Third, the model’s uncertainties and unknown external disturbances are handled by adaptive fuzzy technique. Finally, the effectiveness of the new control scheme is illustrated by the simulation.