A Semantic SLAM Model for Autonomous Mobile Robots Using Content Based Image Retrieval Techniques

  • Authors:

  • Choon Ling Tan;Simon Egerton;Velappa Ganapathy

  • Affiliations:

  • Monash University, Jalan Lagoon Selatan, Selangor Darul Ehsan, Malaysia 46150;Monash University, Jalan Lagoon Selatan, Selangor Darul Ehsan, Malaysia 46150;Monash University, Jalan Lagoon Selatan, Selangor Darul Ehsan, Malaysia 46150

  • Venue:
  • ICONIP '09 Proceedings of the 16th International Conference on Neural Information Processing: Part II
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Semantic approaches to conducting SLAM are still considered to be comparatively new compared to other methods. To this end, we introduce a new model for conducting SLAM on an autonomous mobile robot equipped with vision sensors. Our model consists of four separate stages, each with a specific goal at hand, namely: feature extraction, classification and storage, semantic analysis, and location resolving. This is the first time SLAM has been examined in this way and a set of planned experiments and benchmarks are also discussed which apply the proposed model to environments which are unknown and vary in their structure. Initial experiments are also included where images captured in different indoor locations are shown, along with the similarity scores of these images. Future work and experiments that are intended to be completed are then discussed.