: . 小型微型计算00130) 2(天津大学 电气自动化与信息工程学院,天津 300072) E-mail:******@ 摘 要:传统的视觉同时定位与地图创建(SLAM)依赖于点特征来估计相机位姿。然而在室内环境中存在大量低纹理区域, 使得提取足够多的点特征变得困难。此外,当相机抖动剧烈或转向过快时,基于点特征的 SLAM 系统也并不鲁棒。针对 上述问题,本文提出了一种基于 RGB-D 的点线特征融合 SLAM 算法,利用点特征和线特征的优点,在困难环境下获得了 鲁棒的结果。首先,提出了一种基于特征丰富度的特征提取策略。解决在模糊和低纹理区域内提取特征困难的问题。其次, 设计了一种点线特征关联图,优化线特征匹配效果。该方法不仅参考了线特征之间的相似关系,还考虑了点线特征之间的 几何关系。最后,在构建光束法平差的成本函数时建立自适应模型,实现点线双模态特征的"无缝融合"。本文分别在两个 公开数据集和室内真实场景中进行了算法评估,并与其他先进算法对比。结果表明本文提出的算法具有更好的整体性能。 关键词:机器视觉;同时定位与地图创建;点线特征;自适应模型;低纹理 中图分类号:TP391 文献标识码:A Point-line Feature Adaptive Fusion Indoor SLAM Algorithm LIU Shao-zhe1,LIU Zuo-jun1,HU Chao-fang2,CHEN Hai-yong1 1(School of Artificial Intelligence, Hebei University of Technology, Tianjin 300130, China;) 2(School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China) Abstract: Traditional visual simultaneous localization and mapping (SLAM) depends on point features to estimate camera pose. However, a large number of low texture areas exist in the indoor environment, which makes it difficult to extract enough point features. In addition, the SLAM system based on point feature is also not robust when the camera jitters sharply or turns too fast. To solve the above problems, this paper proposes a point-line feature fusion SLAM algorithm based on RGB-D. The advantages of point feature and line feature are utilized, and robust results are obtained in difficult environments. Firstly, a feature extraction strategy based on feature richness is proposed. It solves the probl