Navigation plays a critical role in enabling the autonomy of aerial vehicles. Traditionally, aerial systems rely on Global Navigation Satellite Systems (GNSS) and inertial measurement units (IMUs) for navigation. Recent research ad-vancements in Simultaneous Localization and Mapping (SLAM) techniques have encouraged their integration into unmanned aerial systems. By combining these techniques with traditional navigation methods, we enhance safety and accuracy during operations. Cameras offer several advantages, including cost-effectiveness, low-power consumption and minimal physical space requirements, while providing rich navigation information. In this paper, we are introducing and testing some of the most common Visual SLAM and provide an insight on how these algorithms perform without any significant modification on aerial datasets. Relevant data, from camera, IMU and GNSS, have been collected using a fixed wing aircraft and drone flying over rural and urban areas at desirable altitudes. The algorithms' output is compared with GNSS measurements, and their performance is analyzed. Additionally, we examine key components of SLAM, such as loop closure, based on our aerial data.
Visual Navigation: Simultaneous Localization and Mapping on Aerial Vehicles
2024-09-29
2437852 byte
Conference paper
Electronic Resource
English
British Library Online Contents | 2018
|Visual navigation system for small unmanned aerial vehicles
Tema Archive | 2013
|European Patent Office | 2024
|Simultaneous localization and mapping for aerial vehicles: a 3-D sensor-based GAS filter
British Library Online Contents | 2016
|