ABSTRACT
Estimation problems may be dynamic, in which the state vector changes as a function of time, or static in which the state vector is constant in time. Within the overall context of multi-sensor data fusion, Kalman filters provide a classic sequential estimation approach for fusion of kinematic and attribute parameters to characterize the location, velocity and attributes of individual entities. A number of sensors observe location parameters such as azimuth, elevation, range, or range rate and attribute parameters such as radar cross section. The location parameters may be related to the dynamic position and velocity of an entity via observation equations. For each sensor, a data alignment function transforms the “raw” sensor observations into a standard set of units and coordinate reference frame. The observability issue concerns the extent to which it is feasible to determine components of a state vector based on observed data.
