Reduced-Complexity Algorithms for Indoor Map-Aware Localization Systems

The knowledge of environmental maps (i.e., map-awareness)can appreciably improve the accuracy of optimalmethods for position estimation in indoor scenarios. This improvement,however, is achieved at the price of a significant complexityincrease with respect to the case of map-unawareness,specially for large maps. This is mainly due to the fact thatoptimal map-aware estimation algorithms require integratinghighly nonlinear functions or solving nonlinear and nonconvexconstrained optimization problems. In this paper, varioustechniques for reducing the complexity of such estimators aredeveloped. In particular, two novel strategies for restricting thesearch domain of map-aware position estimators are developedand the exploitation of state-of-the-art numerical integration andoptimization methods is investigated; this leads to the developmentof a new family of suboptimal map-aware localizationalgorithms. Our numerical and experimental results evidence thatthe accuracy of these algorithms is very close to that offeredby their optimal counterparts, despite their significantly lowercomputational complexity