Simultaneous Localization and Layout Model Selection in Manhattan Worlds
This addresses robot navigation in structured environments, but appears incremental as it builds on existing Manhattan world assumptions.
The paper tackled the SLAM problem by transforming it into a model selection problem using Manhattan structure, solved via convex optimization over layout structures, and demonstrated it on real-world datasets.
In this paper, we will demonstrate how Manhattan structure can be exploited to transform the Simultaneous Localization and Mapping (SLAM) problem, which is typically solved by a nonlinear optimization over feature positions, into a model selection problem solved by a convex optimization over higher order layout structures, namely walls, floors, and ceilings. Furthermore, we show how our novel formulation leads to an optimization procedure that automatically performs data association and loop closure and which ultimately produces the simplest model of the environment that is consistent with the available measurements. We verify our method on real world data sets collected with various sensing modalities.