Visual inertial lidar odometry and mapping (VI-LOAM) fusion for UAV-based parcel delivery

Dissanayaka, Didula (2022) Visual inertial lidar odometry and mapping (VI-LOAM) fusion for UAV-based parcel delivery. Masters thesis, Memorial University of Newfoundland.

[img] [English] PDF - Accepted Version
Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission.

Download (9MB)


This study presents the design, implementation, and validation of a robust GPSaided multi-sensory odometry and mapping system for an end-to-end UAV-based parcel delivery application. There are two main approaches for UAV navigation, GPS navigation and GPS denied navigation. However, the existing GPS navigation solutions sometimes produce degenerative results due to GPS loss, multipath signals, and spoofing events. On the other hand, GPS denied navigation solutions suffer from inherent drift and sensor degradation scenarios. Additionally, the existing navigation solutions do not comply with UAV safety regulations when possible sensor failures end-to-end navigation requirements is considered. Therefore, this thesis focus on developing a robust UAV navigation system by integrating visual, inertial, and lidar sensors with GPS to overcome the limitations of existing navigation solutions. Three significant contributions are produced in this study. First, a numerical study to evaluate the possibility of incorporating GPS with a GPS denied navigation solution for improved performance and safety regulatory compliance. Second, the development of a novel UAV navigation architecture combining visual, lidar, and inertial sensors that is robust for environmental degradation and aggressive motion. Third, integrating GPS with the novel UAV navigation architecture for improved accuracy. Additionally, this study presents results and a comparison study of the proposed navigation system and state-of-the-art navigation systems for different online benchmark datasets and in-house datasets. Moreover, the proposed GPS-aided UAV navigation system is evaluated against compliance with the associated safety regulations under different GPS scenarios.

Item Type: Thesis (Masters)
Item ID: 15877
Additional Information: Includes bibliographical references (pages xi-xxii)
Keywords: unmanned aerial vehicle (UAV), navigation, simultaneous localization and mapping (SLAM), UAV-based parcel delivery, optimization, visual inertial lidar odometry and mapping (VI-LOAM).
Department(s): Engineering and Applied Science, Faculty of
Date: October 2022
Date Type: Submission
Digital Object Identifier (DOI):
Library of Congress Subject Heading: Drone aircraft--Design and construction; Optical radar; Express service; Cartography; Inertial navigation (Aeronautics)--Design and construction

Actions (login required)

View Item View Item


Downloads per month over the past year

View more statistics