A Laser Data Compensation Algorithm Based on Indoor Depth Map Enhancement

The field of mobile robotics has seen significant growth regarding the use indoor laser mapping technology, but most two-dimensional Light Detection and Ranging (2D LiDAR) can only scan a plane fixed height, it is difficult to obtain information objects below so inaccurate environmental navigation mis-collision problems easily occur. Although three-dimensional (3D) LiDAR also gradually applied, less used in because more expensive requires large amount memory computation. Therefore, data compensation algorithm based on depth map enhancement proposed this paper. Firstly, acquired by camera removed smoothed bilateral filters achieve data, multi-layer projection transformation performed reduce dimension compress into pseudo-laser data. Secondly, are remap according positional relationship between two sensors obstacle. Finally, fused added simultaneous localization (SLAM) front-end matching multi-level fusion. performance multi-sensor fusion before after compared with that existing scheme via simulation kind. experimental results show comprehensive perception effectively improve accuracy building.