Dynamic maps for automated driving and UAVs geofencing

TL;DR

This paper introduces a novel approach to creating dynamic, spatio-temporal maps for autonomous vehicles and UAV geofencing, highlighting its architecture, potential 5G applications, and testing object detection performance.

Contribution

It presents a new architecture for dynamic maps and explores their application in UAV geofencing within 5G services, expanding beyond traditional transportation uses.

Findings

Object detection module performance tested and discussed

Proposed system adaptable for various 5G services

Demonstrated potential for enhanced UAV geofencing

Abstract

The past few years have witnessed a remarkable rise in interest in driver-less cars; and naturally, in parallel, the demand for an accurate and reliable object localization and mapping system is higher than ever. Such a system would have to provide its subscribers with precise information within close range. There have been many previous research works that have explored the different possible approaches to implement such a highly dynamic mapping system in an intelligent transportation system setting, but few have discussed its applicability toward enabling other 5G verticals and services. In this article we start by describing the concept of dynamic maps. We then introduce the approach we took when creating a spatio-temporal dynamic maps system by presenting its architecture and different components. After that, we propose different scenarios where this fairly new and modern technology can be adapted to serve other 5G services, in particular, that of UAV geofencing, and finally, we test the object detection module and discuss the results.