Basic Visual and Motor Agents for Increasingly Complex Behavior Generation on a Mobile Robot

  • Authors:
  • Maria C. Garcia-Alegre;Felicidad Recio

  • Affiliations:
  • Instituto de Automática Industrial (IAI), Consejo Superior de Investigaciones Cientificas (CSIC), 28500 Arganda del Rey, Madrid, Spain. E-mail: maria@iai.csic.es;Instituto de Automática Industrial (IAI), Consejo Superior de Investigaciones Cientificas (CSIC), 28500 Arganda del Rey, Madrid, Spain. E-mail: maria@iai.csic.es

  • Venue:
  • Autonomous Robots - Special issue on autonomous agents
  • Year:
  • 1998

Quantified Score

Hi-index 0.00

Visualization

Abstract

Present work addresses the guidelines that have been followed toconstruct basic behavioral agents for visually guided navigation within theframework of a hierarchical architecture. Visual and motor interactions aredescribed within this generic framework that allows for an incrementaldevelopment of behavior from an initial basis set. Basic locomotion agentsas, Stop&Backward, Avoid, and Forward are implemented by means of fuzzyknowledge bases to deal with the uncertainty and imprecision inherent toreal systems and environments. Basic visual agents as, Saccadic,Find_Contour, and Center are raised under a space-variantrepresentation pursuing an anthropomorphic approach. We illustrate how acomplex behavior results from the combination of lower level agents alwaysconnected to the basic motor agents. The proposed methodology is validatedon a caterpillar mobile robot in navigation tasks directed by an objectdescription.