Research Of Localization And Mapping On Fusion Of Geomagnetic Signal And Lidar

A decision system based on high-precision maps and positioning is currently the most feasible technical route for the autonomous driving industry.Simultaneous Localization And Mapping(SLAM)is the key technology for automatic driving to obtain high-precision maps and positioning.Based on the laser SLAM algorithm,a closed-loop detection algorithm combining geomagnetic sequence search and laser point cloud matching is proposed in this paper.For the traditional laser SLAM in the environment of laser beam interference and in the environment of high structural similarity,the closed-loop misdetection that is easy to occur causes erroneous mapping and positioning.Here,a closed-loop detection algorithm combined with geomagnetic sequence search and laser point cloud matching is used to solve the above problems.By adding a geomagnetic matching algorithm to the closed-loop detection to further filter the initial closed-loop detection pose matrix set,the laser closed-loop detection method is reduced Misdetection is caused by high local similarity,and causes closed-loop misdetection in the disturbing environment,incorrect positioning,and distortions in image building to be corrected.The traditional laser SLAM algorithm is compared with the geomagnetic and laser fusion SLAM algorithm proposed in this paper for positioning and mapping effects.The experimental results show that the algorithm proposed in this paper has a significant improvement in matching speed and accuracy,which can effectively solve laser SLAM closed-loop detection algorithm in the face of closed-loop misdetection in the face of interference,as well as the phenomenon of mapping and positioning distortion.Finally,qualitative experimental analysis is performed on three aspects: algorithm speed,efficiency optimization,and stability.The experimental results show that the closed-loop detection speed is improved by 31%(100 matches).