Joint optimization based on direct sparse stereo visual-inertial odometry
- PDF / 9,387,992 Bytes
- 19 Pages / 595.276 x 790.866 pts Page_size
- 38 Downloads / 256 Views
Joint optimization based on direct sparse stereo visual-inertial odometry Shuhuan Wen1,2
· Yanfang Zhao1 · Hong Zhang2 · Hak Keung Lam3 · Luigi Manfredi4
Received: 8 March 2019 / Accepted: 20 December 2019 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract This paper proposes a novel fusion of an inertial measurement unit (IMU) and stereo camera method based on direct sparse odometry (DSO) and stereo DSO. It jointly optimizes all model parameters within a sliding window, including the inverse depth of all selected pixels and the internal or external camera parameters of all keyframes. The vision part uses a photometric error function that optimizes 3D geometry and camera pose in a combined energy functional. The proposed algorithm uses image blocks to extract neighboring image features and directly forms measurement residuals in the image intensity space. A fixed-baseline stereo camera solves scale drift. IMU information is accumulated between several frames using manifold preintegration and is inserted into the optimization as additional constraints between keyframes. The scale and gravity inserted are incorporated into the stereo visual inertial odometry model and are optimized together with other variables such as poses. The experimental results show that the tracking accuracy and robustness of the proposed method are superior to those of the state-of-the-art fused IMU method. In addition, compared with previous semi-dense direct methods, the proposed method displays a higher reconstruction density and scene recovery. Keywords Direct sparse odometry · IMU pre-integration · Sliding window optimization · 3D reconstruction
1 Introduction Recently, simultaneous localization and mapping (SLAM) has been a popular research topic in robotics, because it is
B
Hong Zhang [email protected] Shuhuan Wen [email protected] Yanfang Zhao [email protected] Hak Keung Lam [email protected] Luigi Manfredi [email protected]
1
Key Lab of Industrial Computer Control Engineering of Hebei Province, Yanshan University, Qinhuangdao, China
2
Department of Computing Science, University of Alberta, Edmonton, AB T6G 2E8, Canada
3
Department of Informatics, King’s College London, 30 Aldwych, London WC2B 4BG, UK
4
Institute for Medical Science and Technology (IMSaT), University of Dundee, Dundee, UK
a fundamental building block for many emerging technologies such as self-driving cars (Urtasun et al. 2012), robotic navigation (Usenko et al. 2016), unmanned aerial vehicles (UAVs), virtual reality (VR), and augmented reality (AR). Pose tracking has attracted significant attention in computer vision. While traditional robotic systems such as self-driving cars have largely relied on LiDAR to actively sense the environment and perform self-localization and mapping, visual SLAM and odometry algorithms have greatly improved in terms of performance. Compared with other sensors, the camera and IMUs are inexpensive, ubiquitous and complementary, and can be combined to work jointly. Visual sensors provide
Data Loading...