A novel control method has been proposed by using the genetic algorithm ( GA ) for nonlinear and complex plants. The proposed control strategy is based on a variable structure control, it overcomes the defects of other adaptive methods such as strong dependence to the system. A GA is used to learn to optimally select integral coefficient C. Simulation results verified the effectiveness of the controller. For position control of Direct Current (DC) motor in practice, this method has good performance and strong robustness, and both dynamic and steady performances were improved.