A New Service Model for Vehicle Sensing Data Collection

Abstract

Vehicles are equipped with more and more sensors like cameras and radar system. These can collect lots of useful data, e.g. for road monitoring, high-definition map updates and for information about accidents or other emergencies. To access and utilize this data, key challenges are the huge amount of data and also the different application needs. We present a new solution to coordinate the vehicle data collection, offering different service models and reducing data uploads by avoiding similar data. We use an architecture with a central coordination scheme to avoid redundant image uploads. It has been successfully evaluated with data from 30 cars driving in Munich, created from a simulator. This includes realistic scenarios with dynamic changes and different services.