Research on 2D Laser SLAM Algorithm Fusion of Semantic Information

Aiming at the problem that the grid map constructed by the traditional 2D laser SLAM algorithm can only represent the state of each region and cannot understand the surrounding environment, a 2D laser SLAM algorithm fusion of semantic information is proposed. First, the Cartographer and the YOLO v3 were used to construct the grid map while object detection of the specified obstacles in the environment, respectively. Then, the objects recognized in the depth map were converted into 3D point cloud data, and the point cloud registration algorithm based on the Iterative Closest Point (ICP) was used to match the two adjacent frames of point cloud data. Next, the minimum and maximum values of the point cloud data in the three dimensions were searched to determine the size of the object. Finally, the corresponding model was searched in the 3D model database and created in the grid map in real-time, thus generating a grid map containing the 3D semantic model. With the guidance of semantic information in the map, the robot can complete the navigation and many other tasks more accurately.