## Learning Topological Maps with Weak Local Odometric Information (1997)

### Cached

### Download Links

- [wilma.cs.brown.edu]
- [www.cs.brown.edu]
- [maroon.cs.queensu.ca]
- DBLP

### Other Repositories/Bibliography

Venue: | IN PROCEEDINGS OF IJCAI-97. IJCAI, INC |

Citations: | 135 - 4 self |

### BibTeX

@INPROCEEDINGS{Shatkay97learningtopological,

author = {Hagit Shatkay and Leslie Pack Kaelbling},

title = {Learning Topological Maps with Weak Local Odometric Information},

booktitle = {IN PROCEEDINGS OF IJCAI-97. IJCAI, INC},

year = {1997},

pages = {920--929},

publisher = {}

}

### Years of Citing Articles

### OpenURL

### Abstract

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.

### Citations

8212 | Maximum likelihood from incomplete data via the EM algorithm
- Dempster, Laird, et al.
- 1977
(Show Context)
Citation Context ...gorithm is a straightforward extension of BaumWelch to deal with the relational information and the factored observation sets. The Baum-Welch algorithm is an expectation-maximization (EM) algorithm [ =-=Dempster et al., 1977-=- ] ; it alternates between the E-step of computing the state-occupation probabilitiessat each time in the sequence given E and the current model , and the M-step ofsnding a new model that maxim... |

4313 | A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition - Rabiner - 1989 |

1259 | On information and sufficiency - Kullback, Leibler |

408 | A robot exploration and mapping strategy based on a semantic hierarchy of spatial representations - Kuipers, Byun - 1991 |

184 | Acting under uncertainty: Discrete bayesian models for mobile robot navigation
- Cassandra, Kaelbling, et al.
- 1996
(Show Context)
Citation Context ...1 Such models have proven particularly useful as a basis for robot navigation in buildings, providing a sound method for localization and planning [ Simmons and Koenig, 1995; Nourbakhsh et al., 1995; =-=Cassandra et al., 1996-=- ] . Much previous work has required that the model be specied manually; this is a tedious process and it is often dicult to obtain correct probabilities. An ultimate goal is for an agent to be able... |

170 | DERVISH an office-navigating robot - Nourbakhsh, Birchfield - 1995 |

131 | Map Learning with Uninterpreted Sensors and Eector - Pierce - 1995 |

99 | Integrating grid-based and topological maps for mobile robot navigation - Thrun, Bucken - 1996 |

95 | Maximum Likelihood Estimation for Multivariate Observation of Markov Sources - Liporace - 1982 |

84 | Learning maps for indoor mobile robot navigation - Thrun - 1998 |

83 | Probabilistic navigation in partially observable environments
- Simmons, Koenig
- 1995
(Show Context)
Citation Context ...iscrete states, observations and possibly actions. 1 Such models have proven particularly useful as a basis for robot navigation in buildings, providing a sound method for localization and planning [ =-=Simmons and Koenig, 1995-=-; Nourbakhsh et al., 1995; Cassandra et al., 1996 ] . Much previous work has required that the model be specied manually; this is a tedious process and it is often dicult to obtain correct probabili... |

81 | Unsupervised learning of probabilistic models for robot navigation - Koenig, Simmons |

65 | Maximum likelihood estimation for multivariate mixture observations of markov chains - Juang, Levinson, et al. - 1986 |

63 |
Maximum likelihood estimation for mixture multivariate stochastic observations of Markov chains
- Juang
- 1985
(Show Context)
Citation Context ...aranteed to provide monotonically increasing convergence of Pr(Ej). Baum-Welch has been proven to be an EM algorithm; it has also been provably extended to real-valued observations [ Liporace, 1982; =-=Juang, 1985-=- ] . Our algorithm introduces an additional matrix, and enforces thesrst two geometric consistency constraints on the M-step, but like the standard Baum-Welch it is still guaranteed to converge to a l... |

61 | Error correction in mobile robot map learning - Engelson, McDermott - 1992 |

51 | Passive distance learning for robot navigation - Koenig, Simmons - 1996 |

20 | Learning dynamics: system identification for perceptually challenged agents - Basye, Dean, et al. - 1995 |

17 |
Dervish: An oce-navigating robot
- Nourbakhsh, Powers, et al.
- 1995
(Show Context)
Citation Context ...ns and possibly actions. 1 Such models have proven particularly useful as a basis for robot navigation in buildings, providing a sound method for localization and planning [ Simmons and Koenig, 1995; =-=Nourbakhsh et al., 1995-=-; Cassandra et al., 1996 ] . Much previous work has required that the model be specied manually; this is a tedious process and it is often dicult to obtain correct probabilities. An ultimate goal is... |

13 | Learning hidden Markov models with geometric information
- Shatkay, Kaelbling
- 1997
(Show Context)
Citation Context ...l matrix, and enforces thesrst two geometric consistency constraints on the M-step, but like the standard Baum-Welch it is still guaranteed to converge to a local maximum of the likelihood function [ =-=Shatkay and Kaelbling, 1997-=- ] . The proof is along the lines of the one presented by Juang et al [ 1986 ] for the standard BaumWelch algorithm, and is beyond the scope of this paper. 4.1 Computing State-Occupation Probabilities... |

9 | et al. Autoclass: A Bayesian classification system - Cheeseman - 1990 |

6 | Error correction in mobile robot map learning - McDermott - 1992 |

1 | likelihood estimation for multivariate mixture observations of Markov chains - Maximum - 1986 |

1 | On information and suciency - Liporace - 1951 |