Fusion&Information Acquisition

There are multiple types of information which can be extracted from sensor actions and which affect the fusion of sensor data. Some information can be anticipated in the form of predicting which information will maximally reduce our uncertainty about a random variable, and some of it is after-the-fact and can be used to change the quality of fusion by, for example, selecting different state estimator process models. The next level of improving fusion is by actively determining which information for a sensor system to obtain and therefore taking a proactive roll in the fusion process, rather than simply performing the best fusion of data that is provided