A Dynamic Map-based Framework for Real-Time Mapping of Vehicles and their Surroundings

Abstract

The great attraction that unmanned vehicles have gained in the past years has marked this century as the era of automation. In particular, automated vehicles have recently become the center of attention of many research activities. All of which have had one goal in common: achieving a complete or partial automation of driving functions, all while ensuring safety and security of other traffic agents. It is with this purpose in mind that the concept of dynamic maps has been introduced to allow vehicles to be aware of their surroundings and have precise knowledge of their environment and all of its components. Aiming for a mapping system that sets out a layered view of a vehicle vicinity ranging from highly-static to highly-dynamic data layers, researchers have taken different approaches. In this paper, we present a system, inspired from the concept of dynamic maps, that collects data from user vehicles and maps them out by location and time, keeping a history of all recorded information. We then evaluate the performance of the system module in charge of localization and tracking of users' devices, and present and discuss the obtained results.