Multi-robot learning using non-deterministic argument games

  • Authors:
  • G. S. Mahalakshmi;T. V. Geetha

  • Affiliations:
  • Department of Computer Science and Engineering, Anna University, Chennai 600025, Tamilnadu, India.;Department of Computer Science and Engineering, Anna University, Chennai 600025, Tamilnadu, India

  • Venue:
  • International Journal of Autonomous and Adaptive Communications Systems
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Multi-robot learning becomes interesting and less complex when the participating robots work in coordination. Both centralised and de-centralised control mechanisms of robots exist in the robot literature. However, the choice of these mechanisms varies with the objective of multi-robot learning. This article presents a de-centralised approach to cooperative knowledge sharing among a team of robots which follow non-deterministic argument gaming. The team of robots navigates over a military area and share ideas related to the objects they perceive during their expedition. The knowledge sharing protocol regulates the entire argumentation scenario. The initial predicates of world knowledge of every robot are represented as ontology. The ontology follows a special structure called Nyaya ontology reference model which takes inspirations from Indian philosophy. The entities share their knowledge by accepting or opposing the proposed arguments referring to the inferences deposited in their knowledge bases.