. This paper addresses the problem of building large-scale geometric maps of indoor environments with mobile robots. It poses the map building problem as a constrained, probabilistic maximum-likelihood estimation problem. It then devises a practical algorithm for generating the most likely map from data, alog with the most likely path taken by the robot. Experimental results in cyclic environments of size up to 80 by 25 meter illustrate the appropriateness of the approach. Keywords: Bayes rule, expectation maximization, mobile robots, navigation, localization, mapping, maximum likelihood estimation, positioning, probabilistic reasoning 1. Introduction Over the last two decades or so, the problem of acquiring maps in indoor environments has received considerable attention in the mobile robotics community. The problem of map building is the problem of determining the location of entities-of-interest (such as: landmarks, obstacles), often relative to a global frame of reference (such as ...

This paper addresses the problem of building large-scale maps of indoor environments with mobile robots. It proposes a statistical approach that phrases the map building problem as a constrained maximum-likelihood estimation problem, for which it devises a practical algorithm. Experimental results in large, cyclic environments illustrate the appropriateness of the approach. 1 Introduction The problem of acquiring maps in large-scale indoor environments has received considerable attention in the mobile robotics community. The problem of map building is the problem determining the location of entities-of-interest(such as: landmarks, obstacles) in a global frame of reference (such as a Cartesian coordinate frame). To build a map of its environment, a robot must know where it is. Since robot motion is inaccurate, the robot must solve a concurrent localization problem, whose difficulty increases with the size of the environment (and specifically with the size of possible cycles therein). T...