Abstract
Keywords - driver assistance, man-machine system, system architecture, aerospace experiences, cooperative systems
Mobility is a fundamental human need and it is far more than the desire to travel. Who is not mobile will be limited in his scope of actions. Mobility in modern social and economic systems is not a luxury but a need for all social strata and economic organizations. This is also reflected in the forecasts for the development of transport: Until 2020, the use of cars will increase by 20% and of trucks by 34%. However, even todays level of mobility creates problems with traffic. If nothing is done, these problems cause losses for the economic system and the quality of life. Travelling by car, the most frequent and flexible mean of transport, produces high costs on lives, time and environment by accidents, congestions and pollution.
Automation and assistance is a well known approach in aerospace to solve these kinds of problem. According to the experiences in this area a huge potential to reduce accidents and congestions can be expected for this technological approach. Within limits, the consumption of energy also can be reduced. However, aerospace experience has also demonstrated disadvantages of automation which should be taken into account when developing assistance for ground vehicles. The more automation takes over tasks which are handled quite well by the driver during most of the time the system behaviour directly competes with the driver´s behaviour. In this case a safe and accepted assistance can only be developed if both man and machine possess an adequate knowledge of the behaviour of their respective counterpart in the man-machine system. This approach will be facilitated by mimic human processing procedures in assistance and automation which is shown in some pilot assistance systems.
Moreover, the different aspects of travelling must not be addressed concerning the different traffic systems and vehicles. From a traveller point of view the three task levels for assistance and automation are navigation, manoeuvring and stabilisation. Assistance in route planning and navigation, in public and private transport must be seamlessly. Manoeuvring and stabilisation support is only necessary if the traffic participant is in charge of a vehicle which can be a car, a truck or even a bicycle. This applies even for walking. Hence, from a model point of view there is a strong relation to the kind of vehicle but the methods work across these boarders.
This view is in the center of the DLR Project "Assistance for Traffic Participants". It comprises a new system architecture concerning the three task levels of driving as well as the human like processing for assistance and automation and presents the implication on the system development. Examples from the different part of this project will be demonstrated.