The quality of wireless communication can vary greatly, especially in highly dynamic environments such as road traffic. So far, it has been difficult to make this communication predictable and reliable. This is where the Fraunhofer ESK solution comes in: The researchers have developed a method for monitoring the quality of service (Quality of Service Monitoring ), which continuously monitors communication between vehicles and important environmental parameters such as topographical conditions. The topographical data are acquired from openly available digital maps; but according to a Fraunhofer spokesperson it also would be feasible to obtain these data from available driver assistance systems that use such data from a backend IT server.
Machine learning algorithms can also be used to predict the communication quality for the next few seconds or minutes (prediction) depending on the situation - the system can react to the new conditions in good time.
A possible application scenario is platooning: several trucks drive in close succession in order to save fuel and improve traffic flow. Up to now, however, the vehicles have had to maintain a large, rigid safety distance. The Fraunhofer ESK solution monitors the communication flow of all common ad-hoc and mobile communication technologies such as 802.11p, LTE and, as soon as they are available, 5G, and predicts how reliable they will be in the next few seconds. If the predicted quality is not sufficient, either a reliable communication channel can be used or the safety distance between the trucks can be increased preventively. The Fraunhofer ESK solution thus gives application developers greater scope for implementing safety-critical networked applications such as platooning and threading assistants as well as driverless transport systems.
The Fraunhofer ESK will present its approach at the ITS World Congress in Copenhagen (September 17 to 21, 2018, Hall C3, Stand 054).
Fraunhofer White Paper discussing this topic: https://www.esk.fraunhofer.de/content/dam/esk/dokumente/Whitepaper-Rapid-Innovation.pdf