To solve the uncertainty of sensor information for intelligent vehicle navigation, a 2-level sensor information fusion method based on both BP neural network and fuzzy neural network was presented. A BP neural network was used to fuse information from multi-ultrasonic sensors so that the uncertainty of the sensors’ information can be decreased and high accuracy of obstacle distance can be obtained. In order to realize preferable decision control of navigation, a fuzzy neural network controller was employed for tracking obstacle avoidance to fuse the obstacle distance and the error information between the vehicle and the marked line. High performance of robustness and flexibility for intelligent vehicle navigation can be achieved with the 2-level information fusion method. Simulation and experiment results verified the effectiveness of the proposed approach.