Enhancing Human-Robot Interaction by Amalgamating Spatial and Attribute Based Cognitive Maps of Assistive Robot

Abstract

Assistive robot technology is rapidly developing over the time. The assistive robots are expected to use in many areas such as medical, educational and assistive tasks. Assistive robots are mostly used for caretakers of elderly people. Further it is essential to be succoured by a human friendly robot in order to take care the elderly people in the domestic environments. The assistive robots should be friendly, reliable, active, and comprehensible in order to be a friendly companion for humans. Humans tend to include uncertain terms related to the direction and the distance to describe or express ideas. Therefore the assistive robot should be capable of analyzing and understanding numerical meaning of uncertain terms for the purpose of creating conceptual map for effective navigation in three dimensional space. Furthermore assistive robot should be able to identify the objects based on their attributes and distinguish one object from another object based on that knowledge. Therefore this paper proposes a method to understand the spatial information and object attributes in conversations and constructs a cognitive maps based on that information to improve human robot interaction. The proposed method has been implemented on MIRob platform. The experiments have been carried out in an artificially created domestic environment and the results have been analyzed to identify the behaviors of the proposed concept.