A Visual-Inertial Odometry Using EKF

XIAO-YU WANG, XIAO-PING HU, XIAO-FENG HE

Abstract


Visual-inertial odometry (VIO) is a process of estimating the position and orientation of an mobile platform (e.g., vehicle, robot) using measurements from on-board cameras and IMU sensor. In this paper, a visual-inertial odometry algorithm is presented which can achieves accurate performance. In the proposed method, relative position and orientation between the body frames at different time instants are estimated by MIMU and monoVO separately, then an extended Kalman filter (EKF) is used to combining these estimation for achieving an accurate estimation of the motion. Experiment is conducted to show the effectiveness of the proposed method in real environment.

Keywords


MIMU, Visual-inertial odometry, EKF.Text


DOI
10.12783/dtetr/ecae2018/27758

Full Text:

PDF

Refbacks

  • There are currently no refbacks.