Simultaneous Localization and Mapping(SLAM) has become an essential function of the robot, but existing SLAM algorithms cannot work robustly and stably in a glass environment, especially when using low cost sensors. In this paper, we propose an efficient and robust method to detect glass based on multi-sensor fusion technology of ultrasonic and laser scan data. By integrating the glass detection algorithm with SLAM, we have improved an existing SLAM algorithm to provide more accurate mapping and localization results. On this basis, we proposed a new robot navigation method and experimented on the robot platform, the experiments show that the new navigation frame increased robot navigation efficiency by 11% in glass environment.