The EOC coordinated a large-scale measurement campaign in Braunschweig. In the process, various traffic scenarios were recorded by a large number of sensors on the ground and from the air. The data set thus obtained is used to develop methods with which automated vehicles will be able to determine their position themselves in the future.
As part of the traffic project "KoKoVI" (Coordinated Cooperative Traffic with Distributed Learning Intelligence), the EOC was involved in a large-scale measurement campaign with several DLR institutes in Braunschweig. Two measurement vehicles drove along defined test routes for this purpose. During their journey, the sensors attached to the vehicles recorded the vehicle environment with numerous sensors from the ground perspective. At the same time, they and their surroundings were recorded from a bird's eye view using the EOC's high-resolution 4k camera system on board a helicopter. The measurement campaign made it possible to generate a data set that illuminates different situations in road traffic from as many perspectives as possible and with the help of different sensors. The combination of these data should enable automated vehicles in the future to determine their own position, as well as the surroundings, more precisely.
After an intensive preparation period and various tests of the vehicle and helicopter components in Oberpfaffenhofen and Braunschweig, the one-week measurement campaign of the KoKoVI project started in Braunschweig on 26.09.2022. With the participation of the EOC, the Institute of Traffic Systems Engineering (TS), the Institute of Optical Sensor Systems (OS), the Institute of Communication and Navigation (KN) and the facility Flight Experiments, a variety of traffic-relevant data was collected in Braunschweig and along the surrounding country roads and highways. The 4k camera system on board a helicopter provided a view of the entire traffic situation from above. The high-resolution aerial images show the campaign participants in a variety of traffic situations: Vehicles at high speeds on highways and country roads, in dense inner-city traffic and their interaction with cyclists and pedestrians at complex urban intersections. At the same time, numerous lidar and camera systems recorded data from the perspective of those involved on the ground, so that together with data from stationary measurement systems along the test routes, this resulted in a consistent and rich data set. A focus of the following work of the EOC is the detection and tracking of moving road users in aerial image sequences and the ego-localization of vehicles with AI algorithms.
Within the KoKoVI project, eight DLR institutes and facilities are working together on technologies to network traffic participants in order to meet the future mobility needs of society while making road traffic safer, more efficient and more environmentally friendly. Traffic is viewed as a networked system in which vehicles, cyclists, pedestrians and the traffic infrastructure itself exchange data and cooperate with each other. The resulting added value for people and their mobility will be demonstrated during the project using use cases.
