Simultaneous localisation and mapping algorithm for topological maps with dynamics

N. L. Doh, K. Lee, W. K. Chung, H. Cho

Research output: Contribution to journalArticlepeer-review

9 Citations (Scopus)

Abstract

A simultaneous localisation and mapping (SLAM) algorithm for topological maps is proposed. Especially, this algorithm deals with a semi-permanent dynamics induced by door opening and closing. To deal with the semi-permanent dynamics, nodes are classified into two types. One is an invariant node that is free from the dynamics, and the other is a variant node that is not affected by door opening and closing. Two different approaches are used at the same time: a quasi-static SLAM for the invariant nodes and a dynamic SLAM for the variant nodes. A gross shape of the given environment is represented by the quasi-static SLAM algorithm, and a detailed shape of the environment is revealed via the dynamic SLAM. Experimental results validate that the proposed algorithm produces a topologically consistent map under the semi-permanent dynamics.

Original languageEnglish
Pages (from-to)1249-1260
Number of pages12
JournalIET Control Theory and Applications
Volume3
Issue number9
DOIs
Publication statusPublished - 2009

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Human-Computer Interaction
  • Computer Science Applications
  • Control and Optimization
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Simultaneous localisation and mapping algorithm for topological maps with dynamics'. Together they form a unique fingerprint.

Cite this