Soft robot is an emergent research field which has variant promising applications, and the control of such robots is still challenging. Unlike using different techniques (such as Beam theory, Cosserat theory or high dimensional finite-element method) to model the dynamics of soft robots, this paper introduces a simplified nominal model with uncertainty to describe its dynamic behavior. The link between this simple model and the finite-element method has been established, and a robust controller is proposed, by compensating the uncertainty which is estimated in a finite time by applying different types of estimators. The experiments have been made for different scenarios, and the corresponding results show the efficiency of the proposed method.