Abstract
Abstract. The number of unmanned aerial vehicles (UAVs) is increasing since low-cost airborne systems are available for a wide range of users. The outdoor navigation of such vehicles is mostly based on global navigation satellite system (GNSS) methods to gain the vehicles trajectory. The drawback of satellite-based navigation are failures caused by occlusions and multi-path interferences. Beside this, local image-based solutions like Simultaneous Localization and Mapping (SLAM) and Visual Odometry (VO) can e.g. be used to support the GNSS solution by closing trajectory gaps but are computationally expensive. However, if the trajectory estimation is interrupted or not available a re-localization is mandatory. In this paper we will provide a novel method for a GNSS-free and fast image-based pose regression in a known area by utilizing a small convolutional neural network (CNN). With on-board processing in mind, we employ a lightweight CNN called SqueezeNet and use transfer learning to adapt the network to pose regression. Our experiments show promising results for GNSS-free and fast localization.
Highlights
The last years show an increasing number of unmanned aerial vehicles (UAVs) operating in industrial, military and private areas
Research in the fields of navigation, data acquisition, path planning or obstacle avoidance for UAVs is increasing. Most of these systems need reliable navigation solutions which are mostly based on global navigation satellite system (GNSS) methods and often combined with alternative methods such as inertial navigation systems (INS)
A small convolutional neural network (CNN) which is efficient in terms of processing while maintaining a satisfying accuracy and generalization is desirable. For this purpose we introduce a CNN-based solution for the navigation or localization of a UAV in a known area by using a variant of SqueezeNet (Iandola et al, 2016), which is originally a CNN for classification and achieves AlexNet Krizhevsky et al (2012) accuracy with 50x less parameters
Summary
The last years show an increasing number of UAVs operating in industrial, military and private areas. Research in the fields of navigation, data acquisition, path planning or obstacle avoidance for UAVs is increasing. Most of these systems need reliable navigation solutions which are mostly based on GNSS methods and often combined with alternative methods such as inertial navigation systems (INS). As failures due to gaps in signal coverage caused by occlusions or multipath effects weaken satellite-based navigation, alternative methods for a reliable navigation of unmanned aerial systems (UAS) are necessary
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
