This study presents a decentralised intelligent double integral sliding-mode control (IDISMC) system, which consists of five IDISMCs, to regulate and stabilise a fully suspended five-degree-of-freedom (DOF) active magnetic bearing (AMB) system. The system structure and drive system with differential driving mode (DDM) are introduced first. Then, the decoupled dynamic model of the five-DOF AMB is analysed for the design of the decentralised control. Moreover, a decentralised integral sliding-mode control (ISMC) system is designed based on the decoupled dynamic model to control the five-DOF AMB considering the existences of the uncertainties. Furthermore, since the control characteristics of the five-DOF AMB are highly non-linear and time varying, the decentralised IDISMC system is proposed to further improve the control performance of the five-DOF AMB. In each IDISMC, the adopted double integral sliding surface reinforces the control law with the integral (I) control feature. In addition, the control gains of the IDISMC can be adjusted on-line and the system uncertainty can also be observed simultaneously by using of a modified proportional-integral-derivative neural network (MPIDNN) observer. Thus, the proposed IDISMC combines the merits of the ISMC, adaptive control and neural network (NN). Finally, the experimental results illustrate the validities of the proposed control systems using various operating conditions.