A novel, effective, and feasible attitude-orbit cooperative control algorithm with adaptive sliding mode control and neural network is developed in this study. The algorithm is analyzed and investigated for the problem of the attitude disturbance caused by orbit transfer of small satellites with chemical propulsion. The changing rule of satellite fuel consumption rate and moment of inertia are given by experimental measurement. The interference torque model based on the chemical propulsion and the 6-DOF dynamic model for the mass variation are established respectively, and moment of inertia variation of the satellite is given. A novel adaptive sliding mode controller for the model is designed based on radial basis function (RBF) neural network, which is utilized to approximate the coupling torque of the orbital transfer and unknown disturbances in the space environment. The new controller provides faster convergence and higher precision control in the system. The simulation results show that the designed controller realizes high-efficiency orbit transfer under limited thrust and control torque conditions. Thus, the proposed method is effective and feasible.