This paper presents a novel control system for underwater vehicle. The underwater vehicle is affected by the inevitable surge and measurement error when working. In order to achieve reliable and quick control characteristic, this paper realize the control system by fuzzy compensation based on Kalman predictor with iterative control response. With this predictor and the control error, the fuzzy controller calculates the control compensation, and then the underwater vehicle completes a fast response control. By the control simulation, the results show that the method is effective and feasible.