Localization of Mobile Robots Incorporating Scene Information in a Hierarchical Model

• The success of mobile robots, specially those operating in human environments, relies on the ability to understand human structures. The aim of this work is to develop a localization framework that considers different scenes and a hierarchical model of the environment. A probabilistic model for recognizing scenes including the objects in the environment is implemented. An efficient hierarchical model formed by different topological representations is used to localize the robot. The localization result is improved by semantic scene information. The experiments developed give preliminary results of our framework working in real environments. They uphold the usefulness of integrating the understanding of the environment in a localization process.

Subject headings

NATURVETENSKAP  -- Data- och informationsvetenskap (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences (hsv//eng)

Keyword

Hierarchical models; Place recognition; Robot Localization; Semantic localization

Publication and Content Type

ref (subject category)

kon (subject category)