Overview
Driver assistance systems offer a considerable potential for improving driving safety and comfort. In designing systems to realise this potential successfully, several important issues concerning the human-machine interaction arise and need to be considered. If a driver uses an assistance system that intervenes directly in the driving process, then this intervention constitutes a fundamental change in the driving task. A part of the driving responsibility is delegated to the assistance system, while the driver assumes a more supervisory role. The assistance system is capable of driving the vehicle independently. Hence, depending on the degree of automation, the driver might sometimes begin to feel like a passenger. The quality of the interaction between the driver and the assistance system essentially determines the acceptance of these systems. Driver assistance systems are not the only new features of vehicle operation. Mobile phones, dynamic routing and traffic-state-controlled navigation devices, and internet services are also entering the vehicle. These information and communication systems also need to be operated and thus compete for the attention of the driver. In certain situations, the danger of overburdening drivers or causing excessive distraction could arise. This danger can be reduced effectively by an appropriate design and interconnection of the sub-systems. The project is part of the research initiative INVENT.
The cluster project has three main objectives:
- How should the driving behaviour of intervening assistance systems be designed to achieve the best possible interaction between the driver and the system?
- How should the human-machine interfaces be designed to guarantee intuitive understanding and easy familiarisation with the system?
- What is the impact of the interaction with driver assistance and information systems on traffic safety?
To this end, methodological fundamentals concerning the central topics of driver behaviour, system familiarisation, and traffic safety will be developed. These will be translated to design guidelines and solution approaches for human-machine interfaces, and the results will be validated using prototypes from the linked cluster projects STA and VAS.
The project will first obtain a structured overview of the current state of knowledge and identify open questions to be resolved. A targeted analysis and classification of the spectrum of driving tasks and requirements is planned; this will be augmented by experimental studies of specific driver reactions to particular types of situations. A (re-)searchable and extendable driver behaviour data base will be created forming the basis for assessment of the potential of driver assistance systems and traffic safety. The learning behaviour of the driver is being investigated experimentally in simulator studies on the basis of test systems. The experience and theoretical knowledge gained in these investigations will result in a learning model for operation of technical systems. Design guidelines will be derived for self-explanatory driver assistance and information systems. The standardised evaluation method is specified to be applicable both in a driving simulator and in field tests. It will be investigated how assistance and information systems affect the occurrence or prevention of driving errors. Taking driver behaviour data into account, a model of the relationship between the situational context, the cognitive burden, and traffic safety will be created and tested in driving simulators. These building blocks will then be combined into an evaluation procedure for traffic safety.