Statistical approaches to fusion with uncertainty

Reasoning with uncertainty is a field with many different approaches and viewpoints, with important applications to sensor design and autonomous system development. An attempt is made to unify some of the different approaches by introducing a common philosophical framework under which different calculi may be developed. Each calculus reflects different design choices compatible with the philosophical tenets. The tenets postulate that uncertainty in representations can be viewed as a degree of dispersion of opinions, and that the space of opinions operates as a separate sample space distinct from the underlying sample space on which probabilities are normally defined. Different calculi result when choices are made for the representation of the opinions, the method for combining opinions, the method for juxtaposing multiple sets of opinions, and the way of measuring the spread in the opinions.< >