## Integrating Grid-Based and Topological Maps for Mobile Robot Navigation (1996)

### Cached

### Download Links

Citations: | 96 - 7 self |

### BibTeX

@MISC{Thrun96integratinggrid-based,

author = {Sebastian Thrun and Arno Bücken},

title = {Integrating Grid-Based and Topological Maps for Mobile Robot Navigation},

year = {1996}

}

### Years of Citing Articles

### OpenURL

### Abstract

Research on mobile robot navigation has produced two major paradigms for mapping indoor environments: grid-based and topological. While grid-based methods produce accurate metric maps, their complexity often prohibits efficient planning and problem solving in large-scale indoor environments. Topological maps, on the other hand, can be used much more efficiently, yet accurate and consistent topological maps are considerably difficult to learn in large-scale environments. This paper describes an approach that integrates both paradigms: grid-based and topological. Grid-based maps are learned using artificial neural networks and Bayesian integration. Topological maps are generated on top of the grid-based maps, by partitioning the latter into coherent regions. By combining both paradigms—grid-based and topological—, the approach presented here gains the best of both worlds: accuracy/consistency and efficiency. The paper gives results for autonomously operating a mobile robot equipped with sonar sensors in populated multi-room environments.

### Citations

7052 |
Probabilistic reasoning in intelligent systems: networks of plausible inference
- Pearl
- 1988
(Show Context)
Citation Context ...he probability ���¡¢�¤§¦�� ���¡¢1¦¥¡¢2¦¥�������¥¡¢���¦��¥ ��� which is conditioned on all sensor reading. A straightforward approach to estimating this quantity is to apply Bayes’ rule (Moravec 1988; =-=Pearl 1988-=-). To do so, one has to assume independence of the noise in different readings. More specifically, given the true occupancy ¢¤£¦¥¨§�© of a grid cell , the conditional ��must be assumeds��. ��� probabi... |

3697 |
Artificial Intelligence : A Modern Approach
- Russell, Norvig
- 1995
(Show Context)
Citation Context ...id-based map. For every abstract plan generated using the topological map, there exists a corresponding plan in the grid-based map (in other words, the abstraction has the downward solution property (=-=Russell & Norvig 1995-=-)). Conversely, every path that can be found in the grid-based map has an abstract representation which is a admissible plan in the topological map (upward solution property). Notice that although con... |

1988 |
Robot Motion Planning
- Latombe
- 1991
(Show Context)
Citation Context ...d, which measures the subjective belief whether or not the center of the robot can be moved to the center of that cell (i.e., the occupancy map models the configuration space of the robot, see e.g., (=-=Latombe 1991-=-)). This section describes the four major components of our approach to building grid-based maps (see also (Thrun 1993)): (1) sensor interpretation, (2) integration, (3) position estimation, and (4) e... |

516 |
Dynamic Programming and and Markov Processes
- Howard
- 1960
(Show Context)
Citation Context ...cell; The cost for traversing a grid cell is determined by its occupancy value. The minimum-cost path is computed using a modified version of value iteration, a popular dynamic programming algorithm (=-=Howard 1960-=-) (which bears similarities to A* (Nilsson 1982)). In a nutshell, starting at each unexplored grid-cell, value iteration propagates values through the map. After convergence, each value measures the c... |

497 |
An Analysis of Time-dependent Planning
- Dean, Boddy
- 1988
(Show Context)
Citation Context ...e done very efficiently. Figure 5b, sketches the path taken during approximately 15 minutes of autonomous exploration. The value function can, however, be used to generate motion control at any time (=-=Dean & Boddy 1988-=-), long before dynamic programming converges. Value iteration has the nice property that values are generated for all cells in the grid, not just the current robot position. � start robot Figure 5: Au... |

405 | A robot exploration and mapping strategy based on a semantic hierarchy of spatial representations". Robotics urrd .-l~utunurrruu.s Sgstems
- Kuipers, Byun
- 1991
(Show Context)
Citation Context ...example, indicate the presence of an obstacle in the corresponding region of the environment. Topological approaches, such a those described in (Engelson & McDermott 1992; Kortenkamp & Weymouth 1994; =-=Kuipers & Byun 1990-=-; Matarić 1994; Pierce & Kuipers 1994), represent robot environments by graphs. Nodes in such graphs correspond to distinct situations, places, or landmarks (such as doorways). They are connected by a... |

343 |
Sensor fusion in certainty grids for mobile robots,”Computer
- Moravec
- 1988
(Show Context)
Citation Context ...roduced two fundamental paradigms for modeling indoor robot environments: the grid-based (metric) paradigm and the topological paradigm. Gridbased approaches, such as those proposed by Moravec/Elfes (=-=Moravec 1988-=-) and many others, represent environments by evenly-spaced grids. Each grid cell may, for example, indicate the presence of an obstacle in the corresponding region of the environment. Topological appr... |

289 |
Sonar-based real-world mapping and navigation
- Elfes
- 1987
(Show Context)
Citation Context ...urer (Real World Interface, Inc.) as part of the regular navigation software. Grid-Based Maps The metric maps considered here are two-dimensional, discrete occupancy grids, as originally proposed in (=-=Elfes 1987-=-; Moravec 1988) and since implemented successfully in various systems. Each grid-cell ¢¤£¦¥¨§�© in the map has an occupancy value attached, which measures the subjective belief whether or not the cent... |

288 | The dynamic window approach to collision avoidance - Fox, Burgard, et al. - 1997 |

253 | Probabilistic robot navigation in partially observable environments
- Simmons, Koenig
- 1995
(Show Context)
Citation Context ...uracy of position estimation. In fact, sonar sensors can be understood as landmark detectors that indirectly—through the grid-based map—help determine the actual position in the topological map (cf. (=-=Simmons & Koenig 1995-=-)). One of the key empirical results of this research concerns the cost-benefit analysis of topological representations. While grid-based maps yield more accurate control, planning with more abstract ... |

150 | World modeling and position estimation for a mobile robot using ultrasonic ranging
- Crowley
- 1989
(Show Context)
Citation Context ... the robot position—gives a second source of information for estimating the robot’s position. 3. Wall orientation. The third source of information estimates and memorizes the global wall orientation (=-=Crowley 1989-=-; Hinkel & Knieriemen 1988). This approach rests on the restrictive assumption that walls are either parallel or orthogonal to each other, or differ by more than 15 degrees from these canonical wall d... |

150 |
T.Weymouth, “Topological mapping for mobile robots using a combination of sonar and vision sensing
- Kortenkamp
- 1994
(Show Context)
Citation Context ...ds. Each grid cell may, for example, indicate the presence of an obstacle in the corresponding region of the environment. Topological approaches, such a those described in (Engelson & McDermott 1992; =-=Kortenkamp & Weymouth 1994-=-; Kuipers & Byun 1990; Matarić 1994; Pierce & Kuipers 1994), represent robot environments by graphs. Nodes in such graphs correspond to distinct situations, places, or landmarks (such as doorways). Th... |

147 | Interaction and intelligent behavior
- Matarić
- 1994
(Show Context)
Citation Context ... presence of an obstacle in the corresponding region of the environment. Topological approaches, such a those described in (Engelson & McDermott 1992; Kortenkamp & Weymouth 1994; Kuipers & Byun 1990; =-=Matarić 1994-=-; Pierce & Kuipers 1994), represent robot environments by graphs. Nodes in such graphs correspond to distinct situations, places, or landmarks (such as doorways). They are connected by arcs if there e... |

108 | The mobile robot Rhino
- Buhmann, Burgard, et al.
- 1995
(Show Context)
Citation Context ...t path between them. Both approaches to robot mapping exhibit orthogonal strengths and weaknesses. Occupancy grids are considerably easy to construct and to maintain even in large-scale environments (=-=Buhmann et al. 1995-=-; Thrun & Bücken 1996). Since the intrinsic geometry of a grid corresponds directly to the geometry of the environment, the robot’s position within its model can be determined by its position and orie... |

100 | A comparison of position estimation techniques using occupancy grids
- Schiele, Crowley
- 1994
(Show Context)
Citation Context ...al sensor reading, it constructs a “local” map (such as the ones shown in Figure 2). The correlation of the local and the corresponding section of the global map is a measure of their correspondence (=-=Schiele & Crowley 1994-=-). Thus, the correlation—which is a function of the robot position—gives a second source of information for estimating the robot’s position. 3. Wall orientation. The third source of information estima... |

83 | Learning maps for indoor mobile robot navigation
- Thrun, Bücken
- 1996
(Show Context)
Citation Context ...Both approaches to robot mapping exhibit orthogonal strengths and weaknesses. Occupancy grids are considerably easy to construct and to maintain even in large-scale environments (Buhmann et al. 1995; =-=Thrun & Bücken 1996-=-). Since the intrinsic geometry of a grid corresponds directly to the geometry of the environment, the robot’s position within its model can be determined by its position and orientation in the real w... |

61 |
Error correction in mobile robot map learning
- Engelson, McDermott
- 1992
(Show Context)
Citation Context ...nments by evenly-spaced grids. Each grid cell may, for example, indicate the presence of an obstacle in the corresponding region of the environment. Topological approaches, such a those described in (=-=Engelson & McDermott 1992-=-; Kortenkamp & Weymouth 1994; Kuipers & Byun 1990; Matarić 1994; Pierce & Kuipers 1994), represent robot environments by graphs. Nodes in such graphs correspond to distinct situations, places, or land... |

44 |
Concurrent localisation and map building for mobile robots using ultrasonic sensors
- Rencken
- 1993
(Show Context)
Citation Context ...ects on the estimation of the robot position. Identifying and correcting for slippage and drift is therefore imperative for grid-based approaches to robot navigation (Feng,Borenstein, & Everett 1994; =-=Rencken 1993-=-). Figure 4 gives an example that illustrates the importance of position estimation in grid-based robot mapping. In Figure 4a, the position is determined solely based on dead� 23.1 meters 32.2 meters ... |

34 | where am I’’ Sensors and Methods for Autonomous Mobile Robot Positioning. Ann Arbor - Feng, Borenstein, et al. - 1994 |

32 | Learning to explore and build maps
- Pierce, Kuipers
- 1994
(Show Context)
Citation Context ...n obstacle in the corresponding region of the environment. Topological approaches, such a those described in (Engelson & McDermott 1992; Kortenkamp & Weymouth 1994; Kuipers & Byun 1990; Matarić 1994; =-=Pierce & Kuipers 1994-=-), represent robot environments by graphs. Nodes in such graphs correspond to distinct situations, places, or landmarks (such as doorways). They are connected by arcs if there exists a direct path bet... |

31 |
Environment perception with a laser radar in a fast moving robot
- Hinkel, Knieriemen
- 1988
(Show Context)
Citation Context ...ition—gives a second source of information for estimating the robot’s position. 3. Wall orientation. The third source of information estimates and memorizes the global wall orientation (Crowley 1989; =-=Hinkel & Knieriemen 1988-=-). This approach rests on the restrictive assumption that walls are either parallel or orthogonal to each other, or differ by more than 15 degrees from these canonical wall directions. In the beginnin... |

8 | Where am I?" Sensors and methods for autonomous mobile robot positioning - Feng, Borenstein, et al. - 1994 |