A Possibilistic Kalman Filter for the Reduction of the Final Measurement Uncertainty, in Presence of Unknown Systematic Errors

A Kalman filter is a concept that has been in existence for decades now and it widely used numerous areas. It provides prediction of the system states as well uncertainty associated to it. The original can not propagate correct way when variables are distributed normally or there correlation measurements systematic error measurements. For these reasons, have variations filter, most them mathematically based (like one) on theory probability. Some indeed introduce some improvements, but without being completely successful. To deal with problems, more recently, filters also defined using random-fuzzy (RFVs). These capable propagating distributions normal contributions uncertainty, thus providing overall measurement state predictions. In this paper, authors make another step forward, by defining possibilistic which only considers propagates both random reduces predictions compensating unknown residual contributions.