Sensor data fusion is the key to today’s and tomorrow’s systems that allow comprehensive environmental sensing from a set of sensors and, beyond that, identification of specific conditions or scenarios. According to Steinberg et al. (1999) the process of data fusion is defined as follows: “Data fusion is the process of combining data or information to estimate or predict entity states.” The need to use different sensors is motivated by various aspects. Redundancy, error reduction and the use of complementary data for the acquisition and identification of complex scenarios are among the primary goals. The use of the Kalman-Filter also plays an important role, as it serves to reduce errors in real measured values and also to provide estimates for non-measurable system variables. In Figure 1 and Figure 2 different approaches of data fusion are shown with regard to time aspects (Figure 1) and with regard to complementary data provided by different sensor functions (Figure 2). If data from several heterogeneous sensors are to be brought into a common data fusion, various fusion methods can be considered. Probabilistic data association (PDA) is one of these methods and has been shown to be simple and effective in studies of several publications, e.g. (Schamm and Zöllner (2011)). In very complex scenarios, finite set statistics (FISST) methods are also becoming more common (Mahler (2004)). In this context, the possibility of model building plays a particularly important role for sensors, which allows additional information to be provided about the sensors by their relationship to one another. The data-based process model formation promises thereby very much potential. Furthermore, a crucial approach from the system point of view is the use of plug-and-play capable sensors (see section 3.3). This approach simplifies the handling of the system and enables the easy configuration and exchange of sensors. In the following sections we will introduce the basics of sensor-fusion, some specific applications and an abstraction based on events and behavior-based management (Hall and McMullen (2004)).