Maximum likelihood registration for multiple dissimilar sensors

Abstract

A study of the maximum likelihood registration (MLR) algorithm for spatial alignment of multiple, possibly dissimilar (active or passive) sensors is presented. The MLR algorithm is a batch algorithm which outputs estimates of the registration parameters, registered sensor measurements and registered target location estimates, expressed in a common coordinate system. The Cramer-Rao type bound for registration of multiple dissimilar sensors is discussed and some numerical examples for sensor registration are presented in support of the theory.