2P1-F04 ステレオと単眼を用いた屋外動的低特徴環境における高精度ビジュアルオドメトリシステム

概要

論文の詳細を見る
In this research, we propose a state-of-the-art 6-Degrees-Of-Freedom(6DOF) Visual Odometry(VO) system which fused stereo VO with effective feature sampling, VO using normal vector information of a ground plane and monocular VO. Stereo VO is used to do 6DOF motion estimation in untextured dynamic environments. Next, VO using ground plane information is used to improve the accuracy of roll and pitch angle estimation. Furthermore, monocular VO is introduced to solve the problems in stereo processing. Finally, Unscented Kalman Filter is adopted to fuse these VO to accomplish robust and accurate localization. We present and evaluate experimental results for our system over 730 meters run in a challenging outdoor environment and compare it with ground truth. Proposed VO system can localize the robot's position within 3% error in untextured outdoor environment.