Key points are not available for this paper at this time.
Abstract Simultaneous Localization and Mapping (SLAM) is one of the core technologies for mobile robots. With the increasing complexity of the environment, the existing 2D laser SLAM has poor quality due to the interference of dynamic objects. Meanwhile, the single-sensor laser SLAM has the problem of insufficient positioning accuracy in a single-structured environment. To address the above drawbacks, this paper studied a novel SLAM method using vision and laser fusion in an indoor dynamic environment based on the Cartographer algorithm.In more details, the dynamic target removal method based on LiDAR (light detection and ranging) detection used to solve the dynamic object interference.Moreover, a laser and vision fusion localization method was proposed for improving the limitation of a single LiDAR sensor. LiDAR to verify the effectiveness of the proposed method, experiments are conducted in the building scene and the long straight path scene respectively. The experimental results demonstrated that the proposed method effectively removed the interference brought by dynamic objects during the map building and the absolute trajectory root mean square error was reduced by 85.45% on average in the structural single environment. The experimental results further verified that the proposed SLAM method improved the quality of map building and positioning accuracy in complex environments.
Li et al. (Tue,) studied this question.