一种基于点和线的视觉惯性SLAM算法

A Point and Line Based Visual Inertial SLAM Algorithm

传统的视觉惯性SLAM (Simultaneous Localization and Mapping)算法,在弱纹理场景下,往往存在定位精度差甚至失效的问题,本文提出一种基于点和线特征的视觉惯性SLAM算法。该算法以开源的VINS-Mono (Monocular Visual-Inertial Systems)系统为基础,在此基础上增加了线特征,结合点线特征各自独有的特性,并且提出一种新的数据选择策略,减少了因为线特征的加入而增加的计算量,保证了系统的实时性。实验采用开源数据集Euroc,通过与其他开源算法做对比,对本文算法进行评估,实验结果表明了本文算法的有效性。

Traditional visual inertial SLAM (Simultaneous Localization and Mapping) algorithms often have poor localization accuracy or even failure in weak texture scenes. In this paper, a visual inertial SLAM algorithm based on point and line features is proposed. The algorithm is based on the open source VINS-Mono (Monocular Visual-Inertial Systems) system, which adds line features, combines the unique features of point and line features, and proposes a new data selection strategy, which reduces the computational burden caused by line features. The real-time performance of the system is guaranteed. The experiment uses open source data set Euroc and evaluates the algorithm in this paper by comparing it with other open source algorithms. The experimental results show the effectiveness of the algorithm in this paper.