IMDEA Networks
Information Management in Vehicular Networks: Relevance-Aware Networking for Advanced Driver Assistance Systems
Tobias Meuser, Research Assistant, KOM - Multimedia Communications Lab, TU Darmstadt, Darmstadt, Germany
External Presentation (External Speaker)
Advanced Driver Assistance Systems require an enormous amount of sensor data. The data sensed by the vehicle’s onboard sensors is often not sufficient due to physical limitations. Thus, vehicles exchange Floating Car Data with each other to enhance their knowledge. As the quality of the exchanged Floating Car Data varies, the vehicles need to verify the data received by others.
In the first half of this talk, I discuss our approach to handle uncertainty in vehicular networks. To this end, I aggregate the received Floating Car Data considering sensor- and information-specific properties. The vehicles use this aggregated view to improve their decision-making.
There is a strong relationship between the decision-making of a vehicle and the data required by it. If a vehicle does not change its decision when receiving data, the vehicle does not require this specific data item.
In the second half of this talk, I discuss the possibilities to use knowledge of the vehicles’ decision-making in the data distribution process. The distribution process can be improved by transmitting information with low accuracy as well as dropping information entirely.
At the end of this talk, an overview of the vehicular extension of the publicly available “Simonstrator Platform” is provided, which supports the simulation of vehicular networking applications.
About Tobias Meuser
Tobi studied business informatics at University of Hagen and Computer Science at TU Darmstadt. In 2016, Tobi obtained his Master’s degree with the thesis “Data Collection Management: A Dynamic Approach for Map Information Harvesting”. Since the end of 2016, he is working as a research assistant in the research group “Distributed Sensing Systems” at Multimedia Communications Lab (KOM). His research interest is centered around assessing the quality and relevance of Floating Car Data in vehicular networks. Based on these metrics, the communication efficiency in vehicular networks can be improved to provide additional connectivity features to vehicles.
This event will be conducted in English