Learning occupancy grids with forward models

Publisher Website
Google Scholar
Abstract
This paper presents a new way to acquire occupancy grid maps with mobile robots. Virtually all existing oc- cupancy grid mapping algorithms decompose the high- dimensional mapping problem into a collection of one- dimensional problems, where the occupancy of each grid cell is estimated independently of others. This induces con- flicts that can lead to inconsistent maps. This paper shows how to solve the mapping problem in the original, high- dimensional space, thereby maintaining all dependencies between neighboring cells. As a result, maps generated by our approach are often more accurate than those generated using traditional techniques. Our approach relies on a rig- orous statistical formulation of the mapping problem using forward models. It employs the expectation maximization algorithm for estimating maps, and a Laplacian approxi- mation to determine uncertainty.

This publication has 3 references indexed in Scilit: