Topological maps provide a useful abstraction for robotic navigation and planning. Although stochastic maps can theoretically be learned using the Baum-Welch algorithm, without strong prior constraint on the structure of the model it is slow to converge, requires a great deal of data, and is often stuck in local minima. In this paper, we consider a special case of hidden Markov models for robot-navigation environments, in which states are associated with points in a metric configuration space. We assume that the robot has some odometric ability to measure relative transformations between its configurations. Such odometry is typically not precise enough to suffice for building a global map, but it does give valuable local information about relations between adjacent states. We present an extension of the Baum-Welch algorithm that takes advantage of this local odometric information, yielding faster convergence to better solutions with less data.

Stochastic topological models, and hidden Markov models in particular, are a useful tool for robotic navigation and planning. In previ- ous work we have shown how weak odometric data can be used to improve learning topological models, overcoming the common problems of the standard Baum-Welch algorithm. Odometric data typically contain directional information, which imposes two difficulties: First, the cyclicity of the data requires the use of special circular distributions. Second, small errors in the head- ing of the robot result in large displacements in the odometric readings it maintains. The cumu- lative rotational error leads to unreliable odomet- ric readings. In the paper we present solutions to these problems by using a circular distribution and relative coordinate systems. We validate their effectiveness through experimental results from a model-learning application.

In this paper, we illustrate the use of a novel probabilistic framework for document analysis on typical problems of document layout analysis and graphics recognition. Our system uses an explicit descriptive model of the document class to find the most likely interpretation of a scanned document image. In contrast to the traditional pipeline architecture, our system carries out all stages of the analysis with a single inference engine, allowing for an end-to-end propagation of the uncertainty. 1. Introduction In [1], we presented a probabilistic framework for document analysis and recognition that lets us find the most likely interpretation of a scanned document image according to a descriptive model of the document class. In contrast to the traditional pipeline architecture, we carry out all stages of the analysis with a single inference engine, allowing for an end-toend propagation of the uncertainty. The global modeling structure is similar to a stochastic attribute grammar,...