Abstract: A novel system is proposed to simulate the spatial microgravity environment for small and mini experimental objects.The compound scheme adopts horizontal air-bearing and pneumatic cylinder vertical suspending to simulate spatial microgravity environment.The vertical microgravity status is simulated with the scheme of constant tensility.The compound control scheme adopts radial basis function neural network(RBFNN) and sliding model variable structure controller,where RBFNN is employed to approximate and compensate uncertainties of the system and works as a feed-forward compensator to make the tracking error converge fast.The sliding model variable structure control is employed to obtain system robustness by reducing the influence of RBFNN approximation error and random disturbance.The experimental results show that the control system is effective,and it produces good dynamic performance,high robustness and good self-adaptive capacity.