基于激光提取特征地图的移动机器人快速鲁棒定位

Fast and Robust Localization of Mobile Robot Based on Feature Map Extracted by Laser Range Scans

基于地图的定位方法在移动机器人领域得到了广泛的应用,但由于全局地图中存在大量的噪声和冗余信息,定位的精度和效率仍然面临挑战。为此,提出了一种改进的移动机器人定位方法,该方法首先从结构化环境中提取直线的特征,然后,利用高斯函数在先验全局地图中提取其最近点,构建特征地图,最后,采用基于k-d树搜索的ICP算法对移动机器人的位姿进行估计。通过分析在真实环境中得到的实验结果,可知该方法在平移方向和旋转方向的定位精度分别提高到11mm和0.23°。同时,与传统匹配定位方法相比,提出的方法在真实环境中的运行时间减少了近20%。

This paper proposes an improved method of mobile robot localization.The method firstly extracts the features of straight lines from the structured environment.Then,Gaussian function is used to extract its nearest points from the prior global map,and the feature map is constructed.Finally,the ICP algorithm based on K-D tree search was used to estimate the pose of the mobile robot.By analyzing the experimental results obtained in the actual environment,it can be concluded that the positioning accuracy of this method in the translation direction and the rotation direction is improved to 11 mm and0.23°,respectively.At the same time,compared with the traditional matching location method,the running time of the proposed method in the real environment is reduced by nearly 20%.