Multi-robot Route Following Using Omnidirectional Vision and Appearance-Based Representation of the Environment

  • Authors:

  • Luis Payá;Oscar Reinoso;Arturo Gil;Javier Sogorb

  • Affiliations:

  • Departamento de Ingeniería de Sistemas Industriales, Miguel Hernández University, Elche (Alicante), Spain 03202;Departamento de Ingeniería de Sistemas Industriales, Miguel Hernández University, Elche (Alicante), Spain 03202;Departamento de Ingeniería de Sistemas Industriales, Miguel Hernández University, Elche (Alicante), Spain 03202;Departamento de Ingeniería de Sistemas Industriales, Miguel Hernández University, Elche (Alicante), Spain 03202

  • Venue:
  • HAIS '08 Proceedings of the 3rd international workshop on Hybrid Artificial Intelligence Systems
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this work, a framework to carry out multi-robot route following is presented. With this approach, a team of mobile robots can follow a route that a leader robot has previously recorded or is recording at the moment. Each robot has a catadioptric system on it that provides omnidirectional images of the environment. To carry out the localization and control of each robot during route following, we have made use of an appearance-based philosophy in combination with compression techniques in the Fourier Domain and the Principal Components Analysis subspace. These techniques are especially interesting in the case of panoramic images due to their invariance to orientation and robustness to small changes in the environment. Several experiments with a team of mobile robots have been carried out to demonstrate the robustness of the approach in realistic office and laboratory environments.