Today, it is generally assumed that the driver observes the environment and is thus immediately available as a fallback level, can react to requests for intervention, or takes over the vehicle guidance outside the operational area of automation. Previous studies on transfer scenarios, particularly in the context of conditional automation, show that considerable time reserves of at least 7 to 10 seconds must be reserved for the driver to safely take over the driving task. This cannot be guaranteed in more complex traffic scenarios by current and soon-to-be-available vehicle environment sensor technology, via so-called Car2X communication, and cloud services. Significantly lower time reserves must be expected here. Therefore, innovative strategies of driver-vehicle interaction are necessary, which bring the driver back to the driving task faster. The overall objective of the project is the research, development and implementation of a system for detecting the driver's situation awareness in order to optimise transfer scenarios in highly automated driving and to adapt the automation behaviour to the driver's state. An innovative concept for highly automated vehicles will be developed, which will enable even complex traffic situations in mixed urban traffic to be handled more safely and efficiently through the interaction of driver state detection, driver model and adaptive driver-vehicle interaction.