Integration of topological and metric maps for indoor mobile robot path planning and navigation

Zavlangas, PG; Tzafestas, SG

dc.contributor.author	Zavlangas, PG	en
dc.contributor.author	Tzafestas, SG	en
dc.date.accessioned	2014-03-01T01:17:59Z
dc.date.available	2014-03-01T01:17:59Z
dc.date.issued	2002	en
dc.identifier.issn	0302-9743	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/14744
dc.subject	Autonomous Mobile Robot	en
dc.subject	Indoor Environment	en
dc.subject	Levels of Abstraction	en
dc.subject	Mobile Robot	en
dc.subject	Path Planning	en
dc.subject	Simulation Experiment	en
dc.subject	Spatial Information	en
dc.subject.classification	Computer Science, Artificial Intelligence	en
dc.title	Integration of topological and metric maps for indoor mobile robot path planning and navigation	en
heal.type	journalArticle	en
heal.identifier.primary	10.1007/3-540-46014-4_12	en
heal.identifier.secondary	http://dx.doi.org/10.1007/3-540-46014-4_12	en
heal.language	English	en
heal.publicationDate	2002	en
heal.abstract	Autonomous mobile robots need to use spatial information. about the environment in order to effectively plan and execute navigation tasks. The information can be represented at different levels of abstraction, ranging from detailed geometric maps to coarse topological maps. Each level is adequate for some sub-task, but not for others. In this paper, we consider the representation of spatial knowledge at two different levels of abstraction, which are commonly considered in the robotics literature : the geometric level, and the topological level. We propose to represent the environment by local metric maps connected into a topological network. This technique allows us to use maps that are not metrically consistent on the global scale, although they are metrically consistent locally. The structure allows also the combination of abstract global reasoning and precise local geometric computations. Moreover, this structure reflects the typical organization of indoor environments, where rooms and hallways define independent but connected local working spaces. To navigate in the environment, the robot uses the topological information to plan a sequence of sectors to traverse, and uses the metric information in each sector to locally move within the sector and to the next one. The functioning of the proposed system with respect to onmidirectional mobile robots and results of simulated experiments are presented.	en
heal.publisher	SPRINGER-VERLAG BERLIN	en
heal.journalName	METHODS AND APPLICATIONS OF ARTIFICIAL INTELLIGENCE	en
heal.bookName	LECTURE NOTES IN ARTIFICIAL INTELLIGENCE	en
dc.identifier.doi	10.1007/3-540-46014-4_12	en
dc.identifier.isi	ISI:000181051700012	en
dc.identifier.volume	2308	en
dc.identifier.spage	121	en
dc.identifier.epage	130	en