A methodical framework for the design of multi-sensor systems in automotive applications

Modern cars comprise various driver assistance applications which depend on reliable surveillance of the vehiclepsilas environment. Therefore, sensors known from aviation and military - like radar or laser - are combined to multi-sensor systems and employed to provide specific observation data. The actual choice of the single sensors depends on the scope of the application and in turn has a considerable impact on the overall system performance. We investigate multi-sensor systems composed of different sensor technologies and provide a methodical framework for the design of sensor based applications and the evaluation of a multi-sensor system regarding specific application requirements. We use both sensor and application characteristics to model the system and propose a metric that supports basic design decisions and assessment of alternative sensor systems at an early stage of system development. The presented approach is very flexible and could be used for any application that relies on multi-sensor systems.