Report

Marker-based FastSLAM on the AlphaBot2

João Rafael Azevedo Gonçalves; Maria Teresa Ramos Nogueira — 2024

Key information

Authors:

João Rafael Azevedo Gonçalves (João Rafael Azevedo Gonçalves); Maria Teresa Ramos Nogueira (Maria Teresa Ramos Nogueira)

Published in

June 13, 2024

Description

This paper presents the implementation of the FastSLAM algorithm on the AlphaBot2 using a monocular camera and artificial ArUco markers. FastSLAM addresses the SLAM problem by combining particle filters and extended Kalman filters for efficient trajectory and landmark estimation. Our implementation highlights the advantages of FastSLAM as we also investigate the data association challenge by ignoring the unique IDs of ArUco markers, simulating environments with indistinguishable landmarks. Experimental results demonstrate that FastSLAM maintains reliable localization and mapping despite ambiguous data association. These results highlight FastSLAM's strengths and limitations in real-world conditions.

Publication details

Authors in the community:

Keywords

  • Alphabot2
  • ArUco Markers
  • FastSLAM
  • Data Association
  • Monocular Camera

Publication language (ISO code)

eng - English

Rights type:

Only metadata available

Creative Commons license

CC-BY - Attribution