基于点线特征融合的视觉惯性里程计方法研究

Research on Visual-inertial Odometry Method Based on Point-line Feature Fusion

相机和惯性测量单元组成的基于图像点特征的视觉惯性里程计(Visual Inertial Odometry,VIO),广泛应用于移动机器人定位领域,但会面临点特征退化的问题,使其定位精度受到很大影响。因此,本文提出一种基于点线特征融合的VIO方法,并在EuRoC数据集上进行实验。结果表明:该方法不仅定位精度最优,而且线特征提取的时间较低。

Visual Inertial Odometry(VIO),composed of cameras and inertial measurement units based on image point features,is widely used in the positioning field of mobile robots,but it faces the problem of point feature degradation,which greatly affects its positioning accuracy.Therefore,a VIO method based on point-and-line feature fusion is proposed in this paper,and experiments are conducted on the EuRoC dataset.The results show that this method not only has the best positioning accuracy,but also the time of line feature extraction is low.