LVIO-based map generation and pose estimation of an unmanned drone using a monocular camera for an indoor flight

被引：0

作者：

Kim K.-W. ^{[1
]}

Jung T.-K. ^{[1
]}

Choi Y.-D. ^{[1
]}

Jee G.-I. ^{[1
]}

机构：

[1] Department of Electrical and Electronics Engineering, Konkuk University

来源：

Journal of Institute of Control, Robotics and Systems | 2019年 / 25卷 / 06期

关键词：

3D feature point map; IMU; LIDAR; LVIO; Monocular-camera; Pose estimation; Unmanned drone;

D O I：

10.5302/J.ICROS.2019.19.0056

中图分类号：

学科分类号：

摘要：

For autonomous flight of drone, pose estimation is required. Especially, in case of indoor flight, the importance of pose estimation is increased. Because when the drone is flying indoors, GPS is not received and there are many obstacles around. In this paper, we propose a precise pose estimation method through LVIO (Lidar Visual Inertial Odometry) algorithm using IMU, vision sensor and LIDAR. In addition, we generate a 3D feature point map and used the pre-generated map and monocular camera to perform pose estimation for indoor flight of a small drones. Generally, the indoor environment is mostly narrow, we can use only small size drone for indoor flight. Therefore, we can only mount light sensors on the drone for pose estimation such as monocular camera and IMU. Therefore, in this paper, LVIO is used to generate the precise trajectory when generating the map, and the pose is estimated using only pre-generated map and monocular camera for drone flight. In addition, it is confirmed that precise pose estimation result is obtained by using LVIO through experiment, and pose estimation for indoor flight is performed using only map and monocular camera. © ICROS 2019.

引用

页码：498 / 505

页数：7

共 14 条

[1] Zhang Z., Determining the epipolar geometry and its uncertainty: A review, International Journal of Computer Vision, 27, 2, pp. 161-195, (1998)
[2] Kim Y., Lee B., Lee Y.J., Sung S., Design and performance evaluation of Kalman filter based visual-inertial odometry using RGB-D camera, Journal of Institute of Control, Robotics and Systems (In Korean), 24, 8, pp. 750-756
[3] Hu J.S., Chen M.Y., A sliding-window visual-imu odometer based on tri-focal tensor geometry, Robotics and Automation, 2014 IEEE International Conference on IEEE Hong Kong, China, pp. 3963-3968, (2014)
[4] Qin T., Li P., Shen S., Vins-mono: A robust and versatile monocular visual-inertial state estimator, IEEE Transactions on Robotics, 34, 4, pp. 1004-1020, (2018)
[5] Zhang J., Singh S., LOAM: Lidar Odometry and Mapping in Real-time, Robotics: Science and Systems, 2, (2014)
[6] Ye H., Liu M., Lidar and Inertial Fusion for Pose Estimation by Non-Linear Optimization, (2017)
[7] Zhang J., Singh S., Visual-lidar odometry and mapping: Low-drift, robust, and fast, 2015 IEEE International Conference on Robotics and Automation, pp. 2174-2181, (2015)
[8] Calonder M., Lepetit V., Strecha C., Fua P., Brief: Binary robust independent elementary features, European Conference on Computer Vision, pp. 778-792, (2010)
[9] Hartley R., Zisserman A., Multiple View Geometry in Computer Vision, (2003)
[10] Magnusson M., The Three-Dimensional Normal-Distributions Transform: An Efficient Representation for Registration, Surface Analysis, and Loop Detection, (2009)

← 1 2 →