Abstract: Objectives: To achieve accurate shore-based measurement of the ship's position, speed, yaw angle and other key motion parameters when a ship enters the ship chamber of the ship lift, a ship's position and attitude sensing method based on Lidar was developed. Methods: To address the specific challenges of point cloud occlusion and noise interference during the process of ships entering a ship lift, a method was developed with the goal of improving the accuracy and efficiency of point cloud data processing. This method utilizes the k-d tree and bounding box algorithms, along with an adaptive voxel grid length calculation method. It incorporates a feature-based ship position sensing method and an adaptive attitude perception method, enabling point cloud preprocessing as well as ship position and attitude sensing. The system's accuracy was validated using data from the Sanxia No. 5 ship, and the impact of key parameters was analyzed. Furthermore, the system's accuracy was also verified using data from Lanjian 208 and Huajia 8 ships. Results: The results show that when the voxel filter constant is 1000, the neighborhood is 10, and the search angle step is 0.25°, the ship position and attitude sensing method has good robustness and smoothness while ensuring the operation efficiency and accuracy of the system. The average deviation of the longitudinal position of the ship is 0.810m, the average deviation of the speed is 0.030m/s, and the average deviation of the yaw angle is 0.27°. Conclusions: The position and attitude sensing method proposed can meet the requirements of engineering practice, and provide technical support for the development of the navigation decision-making system for ships entering and exiting the ship lift.