The 3D map building of the mobile robot

Abstract

The map building is the prerequisite in the mobile robots' autonomous operation in an unknown environment. Most current mobile robots mainly use the 2D map, and the others which use the 3D map have some shortcomings such as the complex data processing algorithm, large computation and time-consuming and so on. The paper presents a 3D data scanning system that combined with the 2D laser radar and the precise rotation head to capture the information of the unknown environment, and building a coordinate system model for calculating the transform matrixes of each coordinate system. Then the pose of the robot got by the odometer and electronic compass, the improved ICP algorithm and the transform matrixes are combined to process the environment information for getting the 3D point cloud, finally the 3D point cloud is displayed on the OpenGL. At last, the experiment about the 3D map building is gone on in the laboratory corridor and the indoor environment. The experimental results show that this 3D data scanning system has higher efficiency and accuracy, and can better reflect the real situation of the environment.