基于加权SDF地图的二维激光SLAM方法

2D LASER SLAM METHOD BASED ON WEIGHTED SDF MAP

提出一种基于加权SDF地图的二维激光SLAM方法。该方法使用一种加权SDF地图,该地图模型能以低于栅格的精度构建地图,较传统的占据栅格地图有更好的精度。同时采用列文伯格-马夸尔特(Levenberg-Marquardt, L-M)法求解激光SLAM中扫描匹配问题,有效缓解了高斯牛顿法可能会导致算法不收敛和局部近似不够准确的局限性。通过ROS仿真实验验证与对比,所提方法能有效降低定位误差,且具有较好的地图构建效果。

This paper presents a 2 D laser SLAM method based on WSDF map. The method used a WSDF map that could construct map with less than the accuracy of the grid. It has better precision than traditional occupancy grid map. At the same time, we used the Levenberg-Marquardt(L-M) method to solve the scanning matching problem, which effectively alleviated the limitation that the Gauss-Newton method may lead to the algorithm not converging and the inaccurate local approximation. Through the ROS simulation and comparison, the proposed method can effectively reduce the positioning error and has a good mapping effect.