|
|
說明 | 1 online resource (87 pages) |
文字 | text |
無媒介 | computer |
成冊 | online resource |
附註 | Source: Masters Abstracts International, Volume: 82-06 |
| Advisor: Waslander, Steven |
| Thesis (M.A.S.)--University of Toronto (Canada), 2020 |
| Includes bibliographical references |
| Classical visual-inertial fusion relies heavily on manually crafted image processing pipelines, which are prone to failure in situations with rapid motion and texture-less scenes. While end-to-end learning methods show promising results in addressing these limitations, embedding domain knowledge in the form of classical estimation processes within the end-to-end learning architecture has the potential of combining the best of both worlds. In this thesis, we propose the first end-to-end trainable visual-inertial odometry (VIO) algorithm that leverages a robo-centric Extended Kalman Filter (EKF). The EKF propagates states through a known inertial measurement unit (IMU) kinematics model and accepts relative pose measurements and uncertainties from a deep network as updates. The system is fully differentiable and can be trained end-to-end through backpropagation. Our method achieves competitive results among state of the art classical and learning based VIO methods on the KITTI dataset |
| Electronic reproduction. Ann Arbor, Mich. : ProQuest, 2021 |
| Mode of access: World Wide Web |
主題 | Robotics |
| Artificial intelligence |
| Computer engineering |
| Deep learning |
| Localization |
| Robo-centric EKF |
| Supervised deep learning |
| Visual inertial odometry |
| Electronic books. |
| 0771 |
| 0800 |
| 0464 |
ISBN/ISSN | 9798698545880 |