An intelligence sliding mode controller for rotary steering drilling stabilized platform based on nonlinear disturbance observer is presented. Nonlinear disturbance observer which can converge exponentially with suitable design parameters is used to obs erve the uncertain disturbance of stabilized platform under work condition. Sliding mode controller is designed to guarantee the robustness of the closed-loop system. The adaptive rate of switching gain is designed and sign function is replaced by bipolar sigmoid function to weaken chattering. Finally, genetic algorithm is applied to search the optimal controller parameters, including switching function coefficient, switching gain adaptive coefficient, sigmoid function coefficient and observer coefficient. Simulation results show that nonlinear disturbance observer can observe the uncertain disturbance effectively, controller output is decreased and stabilized platform can get optimal control performance and robustness.