Sensor selection and fusion using subjective logic

Abstract

Sensor fusion is the notion of combining the data from two or more sensors in order to enhance performance compared with that of individual sensors. The most common method for fusing sensors is through Bayesian methods. However, these cannot easily take into account unknown uncertainty or imprecision. A relatively new method is subjective logic. Although similar to Dempster-Shafer theory, it is unique in that it allows us to collapse the frame of discernment into a binary frame, thereby reducing the complexity. In this paper, we show two novel methods for employing subjective logic: 1) it can be used for target identification (and we show some examples for surveillance in the airborne environment); 2) given some knowledge about the performance of a suite of sensors, we might be able to select the best sensor for a given task. This is achieved through the use of the expected decision formula.